Background

Degeneration of one or more joints in the body affects almost all persons from the fifth decade of life and there are equally universal clinical signs by the age of 70 years. There is evidence of the problem occurring in the bone records from the distant past as well, and it is a condition found in almost all vertebrates, even dinosaurs, whales, fish and birds. It is probable that the joint structures reach the limit of their regenerative capacity earlier than other body tissues. They certainly operate at the limit of engineering tolerance and circulatory renewal at the best of times. A number of environmental conditions, life circumstances, traumatic events, infections and other diseases and genetic factors might reduce the regenerative capacity even further.

In fact, there is still much uncertainty over the aetiology of osteoarthritis (OA). Wear and tear is a factor in prolonged overuse of a certain joint, but use is generally better than inactivity (long-distance runners have no worse incidence of the disease but even moderate forced inactivity can accelerate the condition). The immediate event is a disturbance in the behaviour of the chondrocyte, normally a very quiescent and isolated cell responsible for maintaining the cartilage of the joint, leading to its mitosis, then an interactive stimulation of neighbouring osteoblasts to produce the bony overgrowth. Whatever provokes the chondroblasts, it is fair, in the absence of hard evidence, to look at the widest systemic evidence for disturbance around the body and the interpretations of such events by earlier observers.

In looking at the wider picture to understand localised OA, one is taking a similar path to that for the more overtly inflammatory arthritic conditions, the diffuse connective tissue diseases: rheumatoid arthritis, Sjögren’s and Behçet’s syndromes, polymyalgia rheumatica, the lupus erythematoses, polymyositis and the like, as well as the spondylitic diseases, ankylosing spondylitis, Reiter’s syndrome and psoriatic arthropathy. These all share a substantial immunological component and their treatment is best taken along with other diseases of this type (see Chapter 8). Nevertheless, even these diseases, inasmuch as the joints are involved, can be approached with some of the following points in mind.

As with skin diseases (see p. 315), the verdict of the ancients on joint diseases was unanimous. Arthritis represented a toxic accumulation at the site. There was probably an appreciation that joints were uniquely vulnerable structures, fairly obvious ‘bottlenecks’ in the circulatory flow; the notion of a ‘toxic log jam’ at such sites seems commonplace, evidenced in part by the topical treatments usually applied.

Counter-irritation

There was, of course, an immediate imperative in joint pain in the past. It quickly affected mobility and threatened survival. It was also easily focused upon. If the problem represented a local toxic accumulation then local measures to improve circulation were the obvious treatment. Applying heat to the joint would most often reduce pain. Applying stronger heating measures could often have a longer term benefit. Thus throughout the world drastic heating measures were applied to affected joints. At the least, hot poultices or baths with mustard, pepper or cayenne would be used as rubefacients (a milder effect can be had with proprietary liniments and embrocations). Indeed there is currently evidence that ginger applied topically¹ or internally² can benefit OA (see also the ginger monograph). For more dramatic effects stronger remedies were used, those in Galenic terms that would be classified as ‘hot in the fourth degree’, that ‘corrode and cause blisters if outwardly applied’ (see Chapter 1).

Blistering in arthritis was a common technique. A very strong mustard or cayenne application might do it and flaying with stinging nettles (see nettle monograph for relevant clinical trials) was favoured in ancient Europe (even birds and some animals will expose themselves to ant bites in certain circumstances), but more corrosive substances such as croton oil and formic acid (found in ant bites and stinging nettle) became favoured as they were isolated. Blistering was used, for example, among the new industrial working classes of 19th century Britain who still favoured their rural herbal traditions and would visit the recently urbanised herbalist because in this way, and without effective social security or system of disability payments, they would be able to get back to work more quickly (they often did not pay until such results were achieved either!). The application could produce a large blister in minutes; lancing this could yield an impressive quantity of fluid. The fact that pain could be instantly reduced and mobility often at least temporarily improved led to the obvious supposition that the blister fluids carried toxins away and that the associated and obvious heating brought in more healing circulation to do the rest. In modern central Europe, blistering applications have sometimes been replaced with precision-depth epidermal puncture devices followed by irritant applications; fluids pass through the puncture holes without blisters having to form. Similar simple devices are also used in traditional Chinese medicine.

Counter-irritation is often in modern accounts explained as resulting from some form of stimulation of nerve receptors leading to reflex analgesia. This does not do justice to the technique or to its therapeutic context; indeed, both rubefacients and blistering agents, expertly applied, are intrinsically comforting sensations. A better speculation might be that counter-irritation is a proinflammatory technique; the increased vasoactivity and stimulation of other inflammatory mediators can be seen as constituting a therapeutic inflammation, doing painlessly what the body itself does with pain, swelling and disability. It is a principle espoused by Samuel Thomson in 19th-century North America (see Chapter 1), that fever and its local counterpart, inflammation, are not the disease themselves but the body’s defence against disease. Like Thomson’s cayenne and Galen’s heating remedies generally, the topical counter-irritations, whether simply warming liniments and embrocations or blistering heating packs, were recognition that the best therapy was to improve upon nature’s defences and even extend them.

In modern practice blistering is rarely acceptable. However, good results can often be had with heating poultices, footbaths or handbaths, using an agent such as powdered mustard. For a poultice, a slurry made with the yellow powder sold to make English mustard can be smeared on a gauze and held over the affected joint under a hot wet flannel pad. Occasional glimpses of the area under the dressing should warn if blistering is impending, but the most common reaction is a mild erythema only. Fifteen to 20 minutes should be enough application. For a hand- or footbath a dessertspoonful of mustard powder is put into a bowl of hot water just large enough to immerse the affected limb; this in turn is best placed in a larger bowl of hot water better to maintain the temperature for up to 20 minutes. It is still possible to obtain cayenne plasters from specialist suppliers, rubber material impregnated with cayenne, and it may be possible to make one’s own substitute. Russian ointment includes cayenne for a particularly robust liniment. Otherwise strong muscle rubs or the Asian product ‘Tiger Balm’ will provide moderate relief for arthritic joints. Most accessibly, many embrocations and liniments are available in pharmacies and elsewhere, mostly based on mentholated ointments with various rubefacient constituents (e.g. eucalyptus and wintergreen oils) added, although these are likely to have only mild benefits.

The above approaches are mostly indicated for the low level of inflammatory activity associated with OA. They may also be indicated in more volatile inflammations of rheumatic disease, but are sometimes not. Gout is another joint condition that is usually contraindicated. If the joint is too inflamed and hot already, these techniques may be aggravatory. Sensitivity to liniments or heat may be a good guide not to proceed further.

Internal treatments

As indicated earlier, a consistent theme in traditional medicine was to remove toxic accumulations from the affected joints. Equally consistent was the view that this was particularly a burden on the kidneys. Many traditional treatments for arthritis were also diuretic (or diuretic depurative) remedies, perceived as helping the body remove toxic waste through the urine. In European medicine herbs such as Apium (celery), Betula (birch), Taraxacum (dandelion) and Filipendula (meadowsweet) were widely used diuretics in the treatment of arthritic disease.

The juxtaposition of kidney function and joint disease was elaborated in the European naturopathic and North American Eclectic traditions in their focus on acid/alkali balance (the kidneys of course are dedicated largely to maintaining electrolyte equilibrium in the body fluids). The concept has permeated folk culture: acids are bad for joints, alkalis good.

It is widely understood that the body has to eliminate acid metabolites and that joint problems are a classic outcome of failure to do this. Under such circumstances it makes sense to reduce acidic foods from the diet; not (and this is one detail causing much confusion) foods that taste acidic such as citrus and other fruits, but foods that leave an acidic residue after digestion, due to a preponderance of sulphates and phosphates. Proteins are the major example of the latter. One guide is to test the acidity of ash after combustion of the food, given that digestion is enzymatically an analogous process; in this test lemons leave an alkaline ash, and cheese, meat and eggs leave an acidic ash.

However, estimates based on the acid or alkaline nature of the mineral ash of the food do not take into account the incomplete intestinal absorption of various nutrients. A useful paper has reviewed the clinical impact of the potential renal acid loads (PRAL) of various foods and elaborated on clinical strategies for the treatment of chronic inflammatory diseases.³

The PRAL of various foods is calculated from their sodium, potassium, calcium, magnesium, phosphorus, chloride and sulphur content, taking into account the known percentage absorption rates for protein (in the case of sulphur) and minerals.⁴ It is expressed as milli-equivalents (mEq) per 100 g. The PRAL content averages for various food groups are shown in Box 9.1.⁴ A positive value means that the food is acidic and a negative value indicates alkalinity.

The most acidic food was parmesan cheese (PRAL 34.2) and the most alkaline food was raisins (PRAL –21.0).⁴ Plant-based beverages such as wine, tea and coffee are generally alkaline if taken without milk. However, some beers and cola drinks were acidic. Mineral water could be quite alkaline (PRAL –1.8) depending on its origin. The most alkaline fruit was the banana (PRAL –5.5), closely followed by apricots (PRAL –4.8). The most alkaline vegetable was spinach (PRAL –14.0), mainly due to its very high calcium content. The least alkaline vegetables were cucumber and asparagus. Processed meats were the most acidic form of meat or fish consumption and lentils and peas were also mildly acidic. Nuts were variable: hazelnuts were alkaline and peanuts and walnuts were acidic. Egg yolks were highly acidic (PRAL 23.4) and chocolate and cake were moderately so.

In the 1930s William Howard Hay, an American doctor, published his theories on health and disease (for example Hay WH. A New Health Era. London: George G Harrap and Co. Ltd; 1935). His system of health became known as the Hay Diet or the Hay System and for a time was very popular. Its influence on naturopathic thinking is still in evidence. Hay realised that foods were either acid or alkaline in nature and that this was not based on their pH before consumption. Instead, in a predecessor to PRAL values, Hay proposed that, since foods are combusted in the body to give energy, it is the pH of the residue of a food after combustion (or ashing) that will indicate its acidity or alkalinity in the body (as noted above). Hay realised that acid foods are those which are rich in protein, since the sulphur content of the protein creates acidity after combustion. In contrast, the alkaline foods are those rich in potassium, sodium and calcium, since these combust to give an alkaline ash.

Although the Hay System was very similar to naturopathy, the original contribution made by Hay was that the modern diet was too acidic and that many modern diseases resulted from this. Hay also developed the concept of food combining, where high protein foods were separated in the diet from high carbohydrate foods. He also proposed that one day of the week should be ‘alkaline’, when only alkaline foods were to be consumed. As noted above, Hay’s classification of alkaline or acidic foods correlates very closely with the PRAL values developed by Remer and Manz.⁴

Using the work of Remer and Manz,⁴ the Hay System can be put on a modern scientific basis. Moreover, excessively acidic foods can be identified and largely avoided and replaced by highly alkaline foods. According to Hay, this will not only result in benefits for OA and other rheumatic conditions, but also many other chronic disorders. Urine pH can be used to measure progress with the diet and adjustments made accordingly. Recent studies indicate that rendering the urine more alkaline will have a sparing effect on calcium excretion. This implies that an alkaline diet may help to prevent osteoporosis, although a recent study suggests this is not the case, except in older men.⁵

A number of other points about the paper of Remer and Manz are worth noting.⁴ Since the body excretes organic acids in proportion to body weight, the more overweight a person is the more acidic their system becomes. If this excess acidity is linked to some of the negative health aspects associated with obesity, then a more alkaline diet could help to counter these negative influences. The fact that the body excretes organic acids also means that the diet needs to be quite alkaline before the urine pH rises above 7.0 into alkalinity. One flaw in the proposed PRAL values is that some organic acids can be consumed in the diet, for example oxalic and tartaric acids, which are not metabolised by the body. These could influence the PRAL values of certain foods (e.g. oranges, grapes and spinach).

The herbal diuretics mentioned above can be understood as usefully complementing a high-alkali diet based on vegetables and fruits in reducing the rate of joint deterioration in many sufferers from OA.

Diuretic herbs may be augmented by inflammatory modulators traditionally used in arthritic disease. There are a number of treatments that have been subjected to systematic review and found to be promising, including gamma-linolenic acid products such as Borago officinalis (borage seed oil), Oenothera biennis (evening primrose oil) and Ribes nigrum (blackcurrant seed oil), capsaicin from Capsicum annuum (cayenne), curcumin from Curcuma longa (turmeric), Tanacetum parthenium (feverfew), Linum usitatissimum (linseed or flaxseed oil) and Boswellia serrata.⁶ Given the wide disparity of treatments and relatively low number of good studies there was understandably a request for confirmatory trials in each case,⁶ some of which have since been conducted. (See also Chapter 2 and the respective monographs.)

Salix (willow) and Populus (poplar) have appreciable levels of anti-inflammatory salicylates. The review referred to above included a double blind trial that one of the authors of this book conducted on a product that contains a mixture of the two. It showed moderate but significant benefit in the relief of chronic arthritic pain in a placebo-controlled study on 82 patients.⁷ In an open prospective cohort study, researchers examined the effect of a standardised aqueous ethanolic willow bark extract on knee and hip symptoms due to gonarthrosis (chronic wear of the cartilage in the knee joint) and coxarthrosis (non-inflammatory degenerative disease of the hip joint). Eighty-eight patients received the equivalent of 120 to 240 mg/day salicin, 40 patients a standard drug treatment as a reference for 6 weeks. As the study progressed there was a trend towards greater improvement in the willow bark group with better tolerability than for the conventional treatment.⁸ A recent systematic review of other studies on Salix concluded that there was moderate evidence of efficacy for the use of ethanolic willow bark extract in low back pain at doses equivalent to 240 mg/day of salicin (see willow bark monograph).⁹

Other traditional anti-inflammatories have also been investigated in modern times: in a recent double blind, placebo-controlled trial, a standardised Boswellia serrata (frankincense) extract was investigated for the treatment of OA of the knee. Seventy-five patients with mild to moderate symptoms were randomised into three equal groups: placebo, low-dose and high-dose extract. Both doses of Boswellia conferred clinically and statistically significant improvements in pain scores and physical ability scores throughout the 90 days of treatment (see also Boswellia monograph).¹⁰ In two randomised controlled clinical trials, a traditional European remedy, comfrey (applied topically), has been found helpful in reducing ankle swelling.^11,12 Harpagophytum (devil’s claw) is a remedy that has attracted a considerable research effort, generally positive though with some poor-quality studies (see devil’s claw monograph).^13,14 A number of alterative remedies are used non-specifically for arthritis. Notable in European tradition is Urtica (nettle), for which there is now some pilot evidence of efficacy available (see also nettle monograph).¹⁵

A Cochrane review of clinical trials for low back pain, usually with arthritic elements, found that there was positive evidence for Harpagophytum procumbens, Salix alba and Capsicum spp. although there were methodological shortcomings in each case.¹⁶

A more recent contender for an arthritic treatment has been standardised preparations of Rosa canina (rosehip). In one double blind, controlled study among 100 patients with OA of the knee or hip, a modest benefit over placebo was detected over 4 months.¹⁷ In another double blind, placebo-controlled crossover trial in 94 patients with the same condition, researchers found that the effect was minimal, but included some reduction of stiffness and symptoms, and reduced use of back-up analgesics.¹⁸ Others are less obvious in their rationale and rely on traditional reputation to support their use; Actaea racemosa (black cohosh), Curcuma longa (turmeric) and Juniperus (juniper) may feature. However, there are now positive results for curcumin from turmeric, especially in high dose or enhanced bioavailability preparations (see turmeric monograph). The efficacy of some of the above together has been established in a double blind controlled clinical trial and evidence for similar combinations exists elsewhere.¹⁹ Zanthoxylum (prickly ash) might be added in cases where cold and poor circulation are identified as factors.

Osteoarthritis

The diarthrodial joint functions to enable smooth articulation of the two adjoining bone ends, at the same time providing both strength and resilience.²⁰ The joint capsule is stiff fibrous tissue to offer structural support, often reinforced by other structures such as collateral ligaments. The inner lining of the joint capsule is the synovial membrane, consisting of an inner thin (1 to 3 cells thick) layer, supported by an outer innervated and vascularised stromal layer. Synovial fluid fills the joint cavity and acts as the lubricant. It is an ultrafiltrate of blood plasma made viscous with hyaluronic acid. The hyaline cartilage covers the articular surfaces and is a highly specialised connective tissue adapted to load bearing and shock absorption. A sparse population of chondrocytes is distributed through the cartilage extracellular matrix, which is mainly type II collagen, proteoglycans (aggrecans), glycoproteins and water.²⁰

Osteoarthritis (OA) has been defined as the failed repair of damage caused by excessive mechanical stress on joint tissues.²⁰ All joint structures are affected, but the major hallmarks are the destruction of articular cartilage and changes in the subchondral bone. Historically OA was called ‘osteoathrosis’, a term implying the absence of inflammation. However, high-sensitive assays (such as for C-reactive protein, CRP) demonstrate that low-grade inflammation is present and that synovial tissue is also involved in the pathology.²¹

OA is the most common joint disorder and the leading cause of disability in the elderly. Among adults 60 years or older the prevalence in the US of symptomatic knee OA is 10% in men and 13% in women.²² Total hip replacements are estimated to increase by seven times from 2005 to 2030.²⁰ Clinically diagnosed OA occurs in more than 50% of adults older than 65 years and in more than 30% aged 45 to 64 years.²³ Widely accepted risk factors are age, obesity, joint injury, genetics (39% to 65% in twin studies), gender, joint misalignment and metabolic disorders.^24–27 However, the robustness of many of these is still debated.

A Canadian study examined the link between primary OA and occupation.²⁸ Agricultural workers showed a significant excess prevalence of OA, with an observed to expected (O/E) ratio of 1.7 in women and 2.3 in men. Linear trends in prevalences between white collar, ‘mixed’ collar and blue collar workers were also significant, with O/E ratios respectively of 1.0, 2.9, and 2.6 in women and 1.0, 1.2, and 1.7 in men. Specific excess prevalence was found in women among housekeepers (O/E 4.4), and in men among unskilled labour workers (O/E 10.3) and truck drivers (O/E 6.7).²⁸

OA results from a complex interaction of mechanical, biochemical, molecular and enzymatic feedback loops.²⁹ The final common pathway is joint tissue destruction resulting from the failure of cells to maintain the homeostatic balance between matrix synthesis and degradation. As OA advances, the dominant catabolic processes lead to progressive joint tissue lesions. The articular cartilage, subchondral bone, synovial membrane and synovial fluid are all affected.

The precise mechanisms behind cartilage degradation are still unclear. Early on there is an increase of water and a decrease of proteoglycans (aggrecans) and type II collagen.³⁰ The predominant enzymes responsible for cartilage matrix degradation in OA are the matrix metalloproteinases (MMPs) and aggrecanases. Aggrecan is a large aggregating proteoglycan that contains chondroitin sulphate and keratan sulphate and is important for the weight-bearing properties of cartilage. A subclass of MMPs known as ADAMTS appears to be activated in OA and break down aggrecan.³⁰ Later, cartilage mineralisation (predominantly calcium pyrophosphate and phosphate) occurs and can accelerate the inflammation.

Synovial inflammation (synovitis) occurs in early OA but can be subclinical. It is possibly induced by cartilage matrix degradation²⁹ and becomes more extensive as OA progresses, with synovial hypertrophy and hyperplasia occurring.³¹ There are increased numbers of immune cells, such as activated B cells and T lymphocytes. In turn, the synovitis may contribute to progression of the cartilage degradation.

The degeneration and erosion of cartilage has recently been challenged as the primary pathological event in OA.²⁹ Subchondral bone is suggested to play a key role: after all the disease was originally called osteoarthritis because of the prominence of the bone reaction. The subchondral bone plate is in direct contact with the cartilage and could influence its degradation. Evidence from humans and animal models has shown that subchondral bone alterations may precede cartilage degeneration. The nutrition of the articular cartilage is provided in part by the vascular bed of the subchondral bone. Early microvascular damage affecting the venous circulation in subchondral bone is found in OA. There is also increasing evidence that bone marrow lesions (BMLs) and bone cysts have an important role in the pathogenesis of knee OA.³² BMLs are strongly associated with radiological progression of knee OA and BML enlargement predicts increased cartilage loss and the reverse.

The common observation that chronic OA patients can experience flare-ups speaks to it being an inflammatory disease. Inflammation seems to be a very early event in OA, perhaps elicited by the initial traumatic injury. Elevated levels of CRP can be observed well before clinical disease.²¹ Inflammation and its triggers directly affect synovial cells (fibroblasts and macrophages), as well as cartilage chondrocytes, causing them to produce cytokines, particularly interleukin (IL)-1beta and later tumour necrosis factor (TNF)-alpha. (The macrophage is a key inflammatory cell in OA.)²¹ OA cartilage produces a large amount of nitric oxide (NO) and reactive oxygen species. This is attributable to increased expression of inducible NO synthase (iNOS). NO reduces the major anabolic processes and increases the catabolic processes.

Eicosanoids, namely leukotrienes (LT) and prostaglandins (PG) are involved, particularly PGE₂ produced from arachidonic acid by cyclo-oxygenase (COX)-2, followed by PG synthase.²⁹ The use of non-steroidal anti-inflammatory drugs (NSAIDs) and COX-2 selective inhibitors has shown that PGE₂ inhibition does not alter the course of progressive OA.³³ LTs produced by 5-lipoxygenase (5-LOX) could be the reason for this, as COX inhibitors might shunt more arachidonic acid into the 5-LOX pathway, paradoxically worsening the progression of OA. LTB₄ potently increases IL-1beta and TFN-alpha from the synovium. Hence dual inhibitors of eicosanoids, acting on both COX and 5-LOX, are desirable.³³ Upstream activation of nuclear factor-kappaB is stimulated by cytokines, excessive mechanical stress and matrix degradation products, and in turn regulates the expression of cytokines, inflammatory mediators (including COX-2) and several matrix degrading enzymes. It may cause chondrocytes to malfunction.³⁴

While the OA definition links it to mechanical stress, predisposition to such stress could be more important. One study found that OA is more widespread in the body than is apparent from clinical studies.³⁵ This is consistent with other data suggesting that OA is a disease that is primarily dependent on systemic predisposition to a particular type of bone response to mechanical stress. Generalised OA is a strong predictor of disease progression.²⁷ One factor in operation here could be advanced glycation endproducts (AGEs). Increased severity of OA correlates with higher cartilage AGE levels.³⁶ AGEs in cartilage trigger AGE receptors (RAGE) on chondrocytes and fibroblast-like synoviocytes causing increased catabolic activity, for example production of cytokines and matrix degrading enzymes, which degrade and breakdown cartilage.³⁷

The source of pain in OA remains enigmatic, especially at the level of the individual patient. Cartilage is aneural; hence it cannot be the tissue that directly generates pain.²⁰ In contrast, subchondral bone, synovium, marginal periosteum, ligaments and the joint capsule are all richly innervated, but rarely can the precise tissue origin of pain be identified in the individual patient. Imaging studies at the knee joint have shown a correlation between pain and both synovitis and subchondral bone changes. The relatively immediate reduction in pain following total knee replacement could be due to the excision of the subchondral plate in this procedure. Grosser pathological changes such as subchondral bone exposure, osteophytes and oedema probably cause pain in highly advanced disease. In addition, alterations of CNS pathways associated with chronic pain states (central sensitisation or ‘wind-up’) have also been identified in OA patients; the neurotransmitter substance P might be involved here.²³ While PGE₂ is generally acknowledged as a major pain mediator, there are indications that leukotrienes can also play a role in pain.³⁸

Growing evidence from epidemiological studies suggests that OA is also linked to primary cardiovascular disease.³⁹ High prevalences of cardiovascular risk factors and vascular co-morbidity have been described in OA. Factors strongly associated include hyperlipidaemia and hypertension. A higher risk of cardiovascular death is associated with widespread OA and one large study found that men with OA in any finger joint were 40% more likely to die from cardiovascular disease.⁴⁰ One study found that the same risk factors that predict for cardiovascular disease, especially those related to coagulation, thrombosis and blood rheology, predict for OA, but with a lower threshold.⁴¹ The study concluded: ‘that there is a hypercoagulable and prothrombotic condition in osteoarthritis, with hypofibrinolysis and indirect evidence of increased fibrin generation’.

One recent review suggested there is mounting evidence that microvascular pathology plays a key role in the initiation and/or progression of OA.⁴² Disruption of microvascular blood flow in subchondral bone may reduce nutrient diffusion to articular cartilage in OA. Ischaemia in subchondral bone due to microthrombi may produce osteocyte death, bone resorption and articular damage in OA. Another slightly earlier review suggested that vascular disease in subchondral bone may accelerate the OA process.⁴⁰ This is either by reducing cartilage nutrition or via direct ischaemic effects on bone, depending if cartilage damage is a primary or secondary inflammatory event. As further evidence, BML formation could be secondary to vascular events. Regardless of what initiates OA (which is more relevant for prevention), vascular disease is suggested to be highly relevant to the progression of OA.

Based on the above information, any rational therapy in OA must recognise the following issues:

The herb Boswellia serrata is a key agent in the rational therapy of OA. A 2010 review noted the following anti-inflammatory effects of Boswellia or boswellic acids from in vitro and in vivo experiments⁴³: inhibition of 5-LOX, but only minor activity on PGE production, downregulation of TNF-alpha by inhibition of NF-kappaB, inhibition of IL-1beta production and inhibition of C3-convertase of the complement system. Particularly active are 11-keto-beta-boswellic acid (KBA) and acetyl-11-keto-beta-boswellic acid (AKBA). The clinical evidence for Boswellia in OA is good. In particular, there are suggestions from some trials that Boswellia treatment might be disease-modifying, rather than just providing symptom control (see Boswellia monograph). This disease-modifying effect should be no surprise given the range of its anti-inflammatory effects relevant to OA.

Another key rational treatment is the proprietary pine bark extract known as Pycnogenol. A double blind, placebo-controlled trial involving 38 knee OA sufferers found 150 mg/day of the proprietary pine bark OPCs (PBO) for 3 months reduced pain and stiffness and improved physical function.⁴⁴ Differences were quite significant clinically, with a 49% reduction in the WOMAC score and a significant drop in conventional painkiller use. A similar design trial, but with 156 patients and PBO at 100 mg/day again showed a 50% reduction in the WOMAC score and a significant drop in oedema.⁴⁵ A follow-up study of this trial found CRP was lowered by 71% by PBO, and fibrinogen by 37%.⁴⁶ Another similar trial in 100 patients receiving 150 mg/day found reduced stiffness and use of other painkillers.⁴⁷ PBO has beneficial effects on the microvasculature, which, in addition to anti-inflammatory activity, could explain its benefits in OA.

The above discussion proposes that microcirculatory and cardiovascular factors may predispose to OA. This was supported by a UK study in female twins that found a strong protective effect on radiographic hip OA for ‘Allium’ consumption (odds ratio 0.70).⁴⁸ Non-citrus fruit was also protective (OR 0.56). Of course, a major Allium species is garlic, well known for its vascular and microvascular effects.

It is hoped that individual strategies will be devised for each patient with OA, taking some of the insights in this and other chapters into account, but essential elements of herbal treatment are as follows:

Gout

Gout is a common arthritis caused by deposition of monosodium urate (MSU) crystals within joints following chronic hyperuricaemia.⁴⁹ It affects 1% to 2% of adults in developed countries where it is the most common inflammatory arthritis in men. Epidemiological studies suggest its incidence is rising. Diet, lifestyle and genetic defects in renal transporters of urate seem to be the main causative factors in primary gout. Gout and hyperuricaemia are associated with hypertension, type 2 diabetes, metabolic syndrome and renal and cardiovascular diseases.⁴⁹ In fact, gout appears to be a risk factor for all cause mortality and cardiovascular mortality and morbidity that is additional to the risk conferred by its association with traditional cardiovascular risk factors.⁵⁰ These co-morbidities can be either a cause or effect of gout. For example, elevated uric acid contributes to the development of essential hypertension.⁵¹

A 2011 review of the risk factors for gout identified 53 relevant studies.⁵² Alcohol consumption, especially beer and spirits, increased the risk of incident gout. Several dietary factors are also implicated, including meat, seafood, sugar sweetened soft drinks and consumption of foods high in fructose (in keeping with gout’s links to metabolic syndrome). Dairy, folate and coffee intake were each associated with a lower incidence of gout.

Many patients with gout present with an acute attack (flare) of gouty arthritis.⁵³ About 20% of patients have urinary tract stones and can develop an interstitial urate nephropathy. When the serum urate levels persistently exceed 6.8 mg/dL (0.4 mmol/L), extracellular fluids become saturated and hyperuricaemia occurs, with an increased risk of MSU precipitation. Although any joint can be affected, the metatarsophalangeal joint of the big toe is the first joint involved in half of all cases.

MSU crystals are potent inducers of inflammation. Within the joint they trigger a local inflammatory reaction, neutrophil recruitment and the production of proinflammatory cytokines.⁵⁴ Uptake of MSU crystals by monocytes involves interaction with components of the innate immune system, including Toll-like receptors (TLRs) and the NALP-3 (NLRP 3) inflammasome complex that drives production of interleukin(IL)-1beta. The inflammatory effects of MSU are IL-1-dependent.

Arthritic disease caused by accumulation of urate crystals at joints provides a particular indication for herbal remedies. There are a number which are claimed to increase elimination of urate from the kidneys, notably Apium (celery), Urtica (nettle) and Betula (birch). There is no doubt that prescriptions based on such herbs appear to ease the symptoms and even help to prevent recurrence. They thus provide a simple and probably safe treatment, especially when combined with a low purine diet (reduced red meat, offal, oily fish, red wine and port), so that urate metabolites are as reduced as far as possible.

Recent research has further informed the dietary and lifestyle advice for patients. Specifically, moderate intake of purine-rich vegetables (asparagus, mushrooms, peas and so on) or protein is not associated with an increased risk of gout.⁵⁵ Hence, these foods can be consumed in moderation. However, the above advice regarding avoidance or reduction of meat, seafood and alcohol has been soundly confirmed.^55,56 Low fat dairy products can be increased, as these appear to be protective,⁵⁷ and there are some promising preliminary studies supporting the time-honoured association of cherries with lowering serum uric acid.^58,59 Weight reduction with daily exercise and a low glycaemic index diet are also important considerations.⁶⁰

The essentials of herbal treatment are as follows:

Repetitive strain injury/carpal tunnel syndrome

According to a 2007 review in The Lancet, repetitive strain injury (RSI) remains a controversial topic.⁶¹ The label is applied to a wide range of specific disorders affecting the upper limbs or neck. The best known is carpal tunnel syndrome (CTS). Others include cubital tunnel syndrome, tendonitis of the wrist or hand, trigger finger, tennis elbow and so on. As the name implies, RSI results from repetitive movements, awkward postures, sustained force and other factors that impact on the normal use of a joint or group of joints. Other descriptive names for the problem include cumulative trauma disorder and occupational overuse syndrome.⁶¹

RSI is relatively common in adults of working age. The overall prevalence is conservatively thought to be between 5% and 10% of adults, but depending on the occupation this can go up to 40%.⁶¹ Statistics show that industrial workers performing repetitive, monotonous movements are at a high risk, but athletes and musicians are also at risk. Long hours at a computer keyboard or using a mouse also can be factors. The occurrence of CTS is quite high, with estimates ranging up to 14% of adults.⁶² RSI is said to cost US industry about $6.5 billion every year.⁶¹

Given the high cost of RSI, it is surprising to learn that, relatively speaking, not much is known about its causes. Stresses at work and psychological distress seem to be contributive factors.⁶¹ However, the factor common to all cases of RSI is overuse of muscle tendon units, causing a reactive inflammation that leads to pain and swelling (oedema) in the local tissues. This can lead to impairment or entrapment of nerves, which adds to the pain and causes other sensations such as parasthesia.⁶¹ In the case of CTS, pressure at the carpal tunnel in the wrist is the most important factor, and being overweight can add to this.⁶² In fact, some experts suggest that CTS is not really a sub-category of RSI, because overuse of the hands is not thought to play a significant role.⁶²

Conventional medical treatment for RSI basically comprises the use of anti-inflammatory drugs, including cortisone injections.⁶¹ For CTS, surgery is often used as a final resort. For non-specific, work-related RSI, immobilisation followed by exercise and manual therapy is often suggested. From the herbal perspective, published clinical studies in RSI are lacking, but clinical experience suggests there is much that can be done.

Herbs for RSI can be recommended on the basis of the known problems associated with the condition, such as inflammation, restricted blood flow, local oedema and nerve entrapment. There are many relevant anti-inflammatory herbs. These include devil’s claw (Harpagophytum procumbens), cat’s claw (Uncaria tomentosa), Boswellia (Boswellia serrata), celery seed (Apium graveolens), turmeric (Curcuma longa), ginger (Zingiber officinale) and willow bark (Salix species). Anti-inflammatory herbs also work well applied topically; the best ones for this purpose are Arnica (Arnica montana) and comfrey (Symphytum species). (See the relevant monographs for more details.)

Other key aspects of herbal treatment for RSI are as follows:

Fibromyalgia syndrome

Fibromyalgia syndrome (FMS) is a disorder of unknown cause characterised by chronic widespread musculoskeletal pain and symptoms such as fatigue, sleep disturbances, gastrointestinal complaints and psychological problems.⁶³ One key diagnostic condition was the presence of multiple tender points on the body (11 of 18 tender points should be present, as defined by the American College of Rheumatology). However, not all patients meet this diagnostic hurdle. This has led to other diagnostic approaches. One such approach uses a widespread pain index in conjunction with a symptom severity scale.⁶⁴ Subclassification of FMS has also been proposed to better inform treatment approaches.⁶⁵

Some clinicians now consider FMS is part of the spectrum of central sensitivity syndrome (CSS).⁶⁴ Clinical entities included under CSS include chronic fatigue syndrome (CFS), irritable bowel syndrome, temporomandibular disorder, idiopathic low back pain, multiple chemical sensitivity (MCS) and interstitial cystitis. Augmented pain and sensory processing in the CNS is suggested as the most reproducible pathogenic feature of these illnesses. This could be due to deficiencies in serotonergic and noradrenergic, but not opioidergic, transmission in the CNS, together with increases in pronociceptive neurotransmitters such as glutamate and substance P.⁶⁶

As alluded to above, there is considerable co-morbidity between FMS, CFS and MCS. For example, a study of a cohort of CFS sufferers found 40.6% met the criteria for MCS and 15.6% met the criteria for fibromyalgia.⁶⁷ Another study found that 70% of tested CFS patients met the criteria for fibromyalgia.⁶⁸ The reverse association is also strong: 58% of a female group with fibromyalgia met the full criteria for CFS compared to 26.1% of a control group; for males the rates were even higher, 80.0% versus 22.2% for controls.⁶⁹ In fact, it has been suggested that FMS and CFS are possibly the same condition.⁷⁰

However, while the literature does support the concept that there is much overlap between CFS and FMS, there are also studies that have been able to clearly differentiate the two conditions (see below). Since CFS and FMS have radically different clinical definitions that form the basis of their respective diagnoses, the possibility exists that they could represent the same underlying condition, but with different aspects of clinical expression. The analogy is somewhat akin to the proverb of the seven blind wise men who sought to explain the nature of the elephant. One felt the trunk of the elephant and concluded it was like a snake, another felt the tail and concluded it was like a rope, and so on.

In this context, an interesting US study examined 646 patients with CFS and/or FMS in terms of the presence or absence of a set of 32 common symptoms.⁷¹ A technique known as latent class analysis demonstrated that essentially patients could be assigned, on the basis of their symptoms, into one of four classes. The mean symptom counts in Classes 1 to 4 were 26±2.0, 20±2.5, 16±2.8 and 11±2.9 respectively. Hence, each class was reasonably discrete from the others. It was found that the presence of CFS was lowest in the more severe class (Class 1) and highest in Class 4. FMS showed the reverse trend. Severity of fatigue was highest in Class 1 and lowest in Class 4. Hence, FMS could represent a more severe form of CFS:

Despite the considerable evidence of the co-morbidity between CFS and FMS reviewed above, and the suggestion that they are not distinct clinical entities, a number of studies have found distinctive differences between them. A well-proven case in point is the observation that only patients fulfilling the definition of FMS (and not those with CFS) have elevated levels of substance P in their cerebrospinal fluid.^72,73 Substance P is a neurotransmitter involved in pain modulation. Higher levels in cerebrospinal fluid may promote peripheral nerve growth and abnormal pain perception. This suggests that more emphasis must be placed on pain management in the treatment of FMS.

Other differences noted in clinical studies include higher nocturnal levels of melatonin in FMS patients compared with CFS and healthy controls,⁷⁴ elevated levels of plasma endothelin-1 in FMS whereas levels in CFS are normal,⁷⁵ and differing protein profiles in cerebrospinal fluid⁷⁶ and urinary electrophoretic profiles between the two disorders.⁷⁷

Some cardiovascular parameters differ between CFS and FMS. For example, haemodynamic instability using a tilt test demonstrated abnormalities for CFS in two studies. Values for FMS patients did not differ from healthy controls.^78,79 The results suggest that more emphasis must be placed on the cardiovascular system in the management of CFS. Patients with FMS often have marked sleep disturbances. In the opinion of one group, sleep disturbances are more marked in FMS than CFS. For CFS patients often the main complaint about sleep is that it is excessive and non-restorative. In contrast, FMS patients almost invariably suffer from less than average time asleep.⁸⁰

There also appear to be marked differences in the neuroendocrine response between CFS and FMS. In particular, with CFS there may be a deficiency of corticotropin releasing hormone (CRH) production, while in FMS there may be excess. Patients with CFS have a mild central adrenal insufficiency, secondary to either a deficiency of CRH or some other central stimulus to the pituitary-adrenal axis.⁸¹ In contrast, a review on hormonal perturbations in FMS asserted:⁸²

These findings and the work of others suggest that an adrenal hyporesponsive state (adrenal depletion) is a distinctly characteristic feature of FMS.

In addition, there appears to be a state of autonomic dysfunction (dysautonomia) in FMS, with sympathetic nervous system hyperactivity.⁸³ Growth hormone levels may be abnormal in FMS and not in CFS.^73,79 Adult growth hormone deficiency is a well-described syndrome with many features reminiscent of fibromyalgia.⁸⁴ It has been postulated that thyroid hormone resistance is also a factor in FMS.⁸⁵ Patients with FMS were less responsive to injections of thyrotropin-releasing hormone in terms of thyrotropin and thyroid hormone production.⁸⁶ However, this could be the result of hypothalamic malfunction.

Like CFS, FMS can have an infectious trigger.⁸⁷ But also like CFS a wide variety of viruses have been associated with the disorder. Viral induced changes in cytokines have been suggested to play a role in fibromyalgia,^88,89 and one study found abnormal cytokine levels,⁹⁰ but not another.⁹¹

Given the similarities between CFS and FMS discussed above, the recommended herbs for FMS are similar to those for CFS outlined in Chapter 8. Specifically, these include:

In addition, issues that are more unique to FMS also require specific attention. Suggested herbal strategies include:

Internal and external applications

Phytotherapy provides two unique approaches to the treatment of skin inflammations. Firstly, there are particular physical and pharmacological properties in plant constituents that possess topical and cosmetic benefits directly on external application,^1,2 and which are discussed further in Chapter 8 and below. Secondly, there are developed medicinal strategies for treating skin problems as manifestations of internal disease, such that most herbal prescriptions are taken orally.³ For both reasons, and in spite of a relative lack of clinical research data, phytotherapy can be recommended as a dermatological strategy. Nevertheless, skin diseases are among the most complex and inconsistent categories in medicine. There are many (including malignant tumours, bullous diseases, alopecia areata and pigmentary disorders) that are probably beyond conservative treatment, others (such as pityriasis rosea and lichen planus) that usually do not need treatment at all and some that are wholly unpredictable. It is unrealistic to claim consistent performance for phytotherapy either. What is offered below is a number of pragmatic options, productively to be considered along with conventional prescription, as well as other complementary approaches, dietary and psychotherapeutic treatments.

A signpost to an enigma?

There are few clear guides to understanding skin diseases. The origin of a proportion can be traced to simple sources, bacterial, viral or fungal infections for example (although a good practitioner would still aim to investigate ultimate causes for these too) or toxic exposures, but the great majority are largely mysterious. Even attempts to distinguish external (exogenous) and endogenous disease are not very helpful. It may look a simple matter that someone has a contact dermatitis to this, that or the other, or that eliminating milk products apparently resolves eczema, but the internal reasons for such sensitivity are left unaddressed. For most skin disease, it is almost impossible to pick out substantive causative factors. Although dermatology is an impressive discipline in terms of differential diagnosis (distinguishing plantar psoriasis from tinea pedis or sarcoidosis from granuloma anulare), there is remarkably little to say about the causes of most skin diseases and the specialty is remarkably bereft of curative strategies. Topical and systemic corticosteroids largely replaced coal-tar products in the 1950s, but these remain the pillars of dermatological treatment, although palliative only.

It is clear that much skin disease is very complex. Notable are those linked to immunological disorders. Psoriasis is specifically an immunological skin disease, but there are many others with greater or lesser association with internal autoimmune disorders: discoid lupus erythematosus, scleroderma, and dermatomyositis. Already in some such cases links have been made between the skin symptoms and events deep in the body (see also Chapter 8). These conditions are also the most inconsistent and refractory; remission may be possible in some individuals but relapses are at least as common and neither change is likely to show much pattern. Clinical experience is that each individual case history is unlike any other and that the landmarks of deterioration, used by practitioners as clues to the construction of a therapeutic strategy, show few common themes.

Although the various skin inflammations diagnosed as eczema are more likely to be considered as local defects (a dermatitis only), it is also certain that these can reflect a wide variety of environmental, dietary and psychological/emotional influences. Again it is hard to avoid the conclusion that the skin manifests deeper disorders; that treatment seems more appropriate directed from the inside out.

Traditional observers of skin disease were almost unanimous in seeing it as a wider disturbance within. Given the paucity of treatment options in conventional dermatology, it may be time to reconsider some of the older strategies.

A suitable case for cleansing

An almost universal view of skin diseases in the past was that they were signals of inner toxicities, accumulations of irritants that the normal eliminatory functions had failed to remove. The variety of possible problems reflected a great variety of toxicities and most developed traditions had a wide range of diagnoses and treatments. The metaphors used were those found elsewhere in these humoral systems. In the meteorological analogies used, toxicity was often equated with ‘damp’, which in turn reflected disorders of the digestive system and liver. The fluctuation in some skin problems was seen to reflect the influence of ‘wind’, the consequence of disturbances in metabolic balance, in the balance of heating and cooling. Those that currently would be classified as allergic eczemas might have been considered ‘dry’ in earlier times, with reference to the gut and respiratory system. Skin disease could also be predominantly ‘hot’ or ‘cold’. In other words, treatments might extend across the whole range of traditional materia medica, depending on details of the indication, mostly through applying remedies that were also seen to be eliminative.

Folk traditions were generally less sophisticated, but the concept of skin disease as a toxicity symptom was consistent. In European traditions, for example, skin problems were treated with ‘blood cleansers’ and ‘blood purifiers’ and the terms ‘depurative’ and ‘dyscratic’ are also derived from this perspective. In China, some remedies were seen as simply good at eliminating poison and were often used as folk treatments for skin diseases. The more acute and severe the skin inflammations, the more robust the remedy used.

Unfortunately there is a gaping lack of modern clinical research for the traditional internal strategies for skin disease. The one notable exception is, however, encouraging.

Two double blind, controlled, clinical trials published by a London team of dermatologists and immunologists in 1992 showed significant efficacy of a herb formulation taken internally. In one study, 40 adult patients with chronic, refractory, widespread atopic dermatitis in crossover between active treatment and placebo herbs showed that substantial benefits followed the use of the Chinese herbs.⁴ In a second, 47 children with a chronic extensive morbilliform variant of non-exudative atopic eczema were given Chinese herbs and placebo herbs in random order, each for 8 weeks, with an intervening 4-week washout period. In 37 who completed the study, the difference in benefits was clear.⁵ Follow-up studies in both cases were also positive. One year after the trial, all subjects on treatment in the first study showed significant benefit; although none were able to discontinue permanently, most had reduced dosage. By contrast, mean scores for untreated individuals showed deterioration.⁶ A 1-year follow-up on the children trial showed that over 50% of patients showed significant lasting benefit and 20% were able to stop treatment with complete remission.⁷

Inevitable safety issues were raised, not least by the authors of the papers themselves (who recommend routine monitoring of liver function tests and tight exclusion criteria for treatment). For example, two cases of acute hepatic illness have been associated with the use of Chinese herbs for eczema⁸ and one case of severe cardiomyopathy was reported after a 2-week course of such treatment.⁹ Nevertheless, actual reports in the clinical trials were not alarming. In the follow-up of the children trial after a year of continuous treatment, two patients taking treatment had raised asymptomatic AST (aspartate aminotransferase) levels, but there were no abnormalities in either full blood counts or biochemical parameters in any adult patient on continued treatment after a year.

The implications of such work have been lost on wider medical thinking, and follow-up attempts to develop a less complex marketable formula foundered. However, it demonstrates that it is possible to design appropriate studies applying traditional diagnoses as inclusion criteria and it is fervently to be hoped that more studies will be forthcoming. One case where internal approaches have been tested is a study on the effects of orally consumed cocoa flavanols on photosensitivity (as measured by skin sensitivity, cutaneous blood flow, transepidermal water loss, and skin structure, texture and hydration after UV irradiation). In a controlled study comparing high and low flavanol cocoa in 24 healthy women over 12 weeks, those consuming the high flavanol version had significantly reduced skin sensitivity to UV.¹⁰ This study reinforces similar protective properties reported for carotenoids and lycopene products. Similar studies exist for the internal use of Centella asiatica (see the gotu kola monograph).

Topical treatments

Chapter 2 will lead the reader to many plant constituents with appreciable direct action on body tissues. There is a wide range of herbal applications to skin lesions, where the skin barrier is damaged. The section on topical applications in Chapter 8 should also be consulted. In summary, the following characteristics of herbal remedies are relevant.

A strategic approach

A good clinician will have no set treatments for skin diseases. It is of course helpful to have a good diagnosis, so that treatment can be better directed. However, even then one reverts quickly to first principles. How are the fundamental body functions? In particular, is there evidence of difficulties in digestive, hepatobiliary and bowel or other eliminatory functions? If there is an immunological component, as in autoimmune or allergic conditions, then events at the gut wall or lumen are even more likely to be factors; they should be tracked down assiduously, with techniques such as rigorous experimental dietary eliminations to elucidate particular difficulties. The priority of treatment is to work at such ‘primary lesions’. It may then be helpful to apply a humoral classification; is there evidence of patterns that might once have been classified as damp, dryness, cold, heat and/or wind? Remedies that dry, moisten, cool, heat or balance respectively could then feature in the prescription. More widely, skin disorders can be considered as inflammatory conditions and approached strategically using insights developed in the relevant sections in Chapter 8.

The treatment of acne and furunculosis may particularly suggest bowel treatments: rosacea and allergic drug eruptions, heptobiliary and bowel remedies. Particularly in such cases one would understandably revert to traditional alterative depurative remedies (see below), especially if their role overlapped with what had been determined as primary problems.

Healing crises?

Skin disease is particularly prone to exacerbations during treatment. It can take very little to provoke this adverse result and ‘healing crises’ are common in homeopathic and dietary treatments of skin disease (for example, in fasting). Some proponents of the latter disciplines claim these are a good thing, a sign that ‘toxins are coming out’. An immersion in pathophysiological mechanisms of skin inflammation would discourage this view and it is not even internally consistent as a notion. Skin inflammation is by definition an extraordinary event, involving a range of traumata in the dermal tissues that have little intrinsic value. If in traditional terms skin diseases suggest inadequacies in the ordinary eliminatory and processing functions, then having even more toxins coming out of the skin, even briefly, does not recommend itself.

There is only one exception to the inadequacy of the healing crisis as a technique in treating skin disease. In cases of low-grade chronic skin trouble, where lack of activity is a characteristic of the condition, promoting subacute or acute crises has been a traditional manoeuvre to render the condition a little more vulnerable to treatment. In most other cases, clinical experience suggests exacerbations beyond a few initial days rarely lead to benefits in the long term; indeed, the opposite is normally true.

A good herbal practitioner will therefore aim for the minimum exacerbation, promoting the defective functions so as to diffuse the pathology. Given the readiness with which exacerbation does occur, a treatment strategy that led to progressive relief without a healing crisis would be something of a triumph.

Atopic dermatitis (eczema)

Eczema, or dermatitis, is a pruritic inflammatory skin reaction that manifests with variable clinical and histological pictures. Atopic dermatitis (AD) is a dermatitis linked to the atopic state. The patient is troubled by itching skin and there is a history of chronic or chronically relapsing dermatitis, worse on the flexures, and a family or personal history of atopy (asthma, hayfever and urticaria). The incidence of AD is increasing worldwide. Patients with AD are at a higher risk for progressing in the atopic march to asthma.¹⁸

AD can be categorised into extrinsic and intrinsic types.¹⁹ Extrinsic or allergic AD shows high total serum IgE levels and the presence of specific IgE for environmental and food allergens, whereas patients with intrinsic or non-allergic AD exhibit normal total IgE values and the absence of specific IgE. While extrinsic AD is the classical type with a high prevalence, the incidence of intrinsic AD is approximately 20%, with a female predominance. Clinical features of intrinsic AD include relatively late onset, milder severity and Dennie-Morgan folds, but no ichthyosis vulgaris or palmar hyperlinearity. The skin barrier is perturbed in the extrinsic, but not the intrinsic, type. Filaggrin gene mutations are also not a feature of intrinsic AD (see below). The intrinsic type is immunologically characterised by the lower expression of interleukin (IL)-4, IL-5 and IL-13, and the higher expression of IFN-gamma. It is suggested that intrinsic AD patients are not sensitised with protein allergens, which induce Th2 responses, but with other antigens, and metals might be one of the candidates for such antigens.¹⁹

As touched on above, specific genetic defects in the epidermal barrier occur in a significant percentage of AD sufferers. The loss-of-function mutations in the structural protein filaggrin, with resultant enhanced transepidermal water loss, are consistent with a unifying hypothesis that offers a mechanistic understanding of AD pathogenesis.²⁰ A diminished epidermal defence to allergens and microbes is followed by polarised Th2 lymphocyte responses with resultant chronic inflammation, including autoimmune mechanisms.

There is a longstanding controversy as to whether allergy is a major pathogenic factor in AD. Several studies have associated food allergy, inhalant allergens and skin contact with airborne allergens with AD.²¹ Even in the absence of a specific IgE for house dust mites, infants with AD have proliferative T cell responses to these antigens. The role of restricting food allergens in the treatment of AD remains controversial in the mainstream literature,²² although one review suggested that egg-exclusion diets could be of value.²³ Another review proposed that, while food allergy is an important provoking cause of AD, elimination diets are only relevant in about 35% of affected individuals.²⁴ Milk, eggs, wheat, soya and peanuts account for 75% of the cases of food-induced AD. Other clinicians do advocate elimination diets over a period of 4 to 6 weeks in AD sufferers with a high serum IgE and positive skin prick tests.²⁵

Other allergens implicated in AD include those from microorganisms and aeroallergens such as pollen, mold and dust mite.²⁵ Dust mite sensitivity appears to be particularly important.²⁶ One clinic with experience of more than 18 000 AD patients also suggested a role for pseudoallergic mechanisms through toxic environmental agents (pollutants, solvents, pesticides and so on).²⁷ A double blind, controlled trial found that house dust mite avoidance measures greatly reduced the activity of AD, especially in children.²⁸ Responses to this therapy varied considerably, despite the fact that allergic reactivity to house dust mite antigens can be established by prick-test challenge in virtually all patients with AD.

These observations highlight a potential misconception concerning allergen exposure in this disorder: a positive skin-prick test (which tests for an IgE-based response) does not necessarily identify those allergens that might be contributing to the underlying immunological disturbance. Some patients with a positive skin reaction to dust mite do not respond to reduced exposure. A corollary is that food allergens are not necessarily identified by a prick-test challenge, although the study described below illustrates that it can be a useful, but not infallible, guide.

Twenty-six children with AD and markedly elevated serum IgE concentrations were evaluated for clinical evidence of hypersensitivity to foods with double blind, placebo-controlled food challenges. Selection of foods for challenges was based on positive skin prick tests (>3 mm wheal) or a convincing history. At least one positive skin test to a food antigen was found in 24/26 patients. A total of 111 double blind, placebo-controlled challenges were performed in these children after suspect foods were eliminated from their diets for 10 to 14 days. There were 23 positive challenges in 15 children, 21 of which manifested as cutaneous symptoms, primarily pruritus and an erythematous macular and/or maculopapular rash involving 5% (or greater) of the body surface. In all, 14 children (54%) developed cutaneous symptoms after food challenges. All symptoms occurred within 10 minutes to 2 hours of challenge; nasal symptoms, mild wheezing and gastrointestinal symptoms were seen in some children. No symptoms occurred in 104 placebo challenges. There were 86/111 clinically insignificant positive skin tests (77%) and three false-negative skin tests. These studies demonstrate that, in some children with AD, immediate food hypersensitivity can provoke cutaneous pruritus and erythema, which leads to scratching and subsequent eczematous lesions. Foods that commonly elicited symptoms on challenge were milk, wheat, eggs, soya and peanuts.²⁹ (This study highlights that replacing cow’s milk with soya milk is not advisable in some patients.) In a follow-up study, elevated plasma histamine levels were found in the group of subjects who had positive reactions to food challenges.³⁰ Other studies have found that chocolate, seafood, oranges, celery and yeast can also commonly provoke symptoms in patients with AD.^31,32

Many types of inflammatory cells are present in AD skin lesions, but the major abnormality is thought to involve hyperstimulatory T cells.²¹ Much interest has focused on the shift in T-helper-cell activity towards a Th2 type response. Both Th1 and Th2 cells can induce B cells to produce immunoglobulins, but only Th2 cells induce IgE.²¹ Regulatory mechanisms are also relevant, with an important role for T regulatory cells.³³

Innate immune responses are also defective in AD, leading to increased susceptibility to viral, bacterial and fungal infections.³⁴ An example of this is the almost 90% rate of colonisation of the skin with Staphylococcus aureus in AD patients, compared with only 5% colonisation in healthy people. Circulating NK cells are significantly reduced in AD patients and are functionally defective, as noted by the reduced release of IFN-gamma. A striking finding in lesional biopsies from AD patients is the absence of neutrophils.³⁵

Although the Th1/Th2 paradigm, with Th2 responses predominating in AD, has been helpful, the actual processes involved are more complex. Patients appear to develop a biphasic helper T-cell pattern, with Th2 cytokine predominance seen early in the acute stage, but then a switch to a more Th1-like profile, with higher levels of IFN-gamma in the chronic stage.³⁴

As noted, the cutaneous microbial flora of atopic dermatitis patients shows striking differences in terms of the presence of Staph. aureus. The relative rarity of colonisation on normal skin is in sharp contrast to the high rate found in patients with AD, ranging from 76% on unaffected areas up to 100% on acute, weeping lesions.³⁶Staph. aureus can induce inflammatory reactions via a range of activities, including toxin and protein secretion. Among these are the superantigens, which have potent inflammatory and immunological effects.³⁶ (Superantigens bind directly to macrophages without antigen processing. This can have profound pathological effects due to the release of cytokines by these cells or via the subsequent activation of T-cells.^37,38) The superantigen Staph. aureus enterotoxin B induces the expansion of Th2 cells, leading to increased IgE synthesis.³⁶

An interesting prospective study comparing, 110 AD patients and 30 healthy volunteers (age 11 to 45 years) demonstrated a high colonisation density of skin lesions and nasal, pharyngeal and vaginal mucosa with Staph. aureus in 102 cases, with streptococci in 53 cases and with Candida, Aspergillus or Penicillium sp. in 36 cases. Quantitative investigations of faecal and duodenal aspirate microflora in the same AD group revealed significantly increased counts of haemolytic coliforms, Candida/Geotrichum and pathogenic Clostridia, generally associated with dramatically reduced counts of lactic acid producing bacteria. By contrast, positive skin cultures with Staph. aureus were isolated in only two controls and increased Candida counts in faeces were found in another three. Specific IgE antibodies against Candida albicans, Aspergillus fumigatus and Saccharomyces cerevisiae were evident in 61, 32 and 56 cases, respectively, suggesting an increased infectious susceptibility and sensitivity to fungal antigens in the disease group. Thirty-one of 58 tested AD sera showed obviously decreased gamma-globulin levels (IgG and IgM, p<0.005) and in 24 of 35 patients tested for delayed cutaneous hypersensitivity reactions, a severe depression of the cellular immune response was recorded. The authors suggested that their experience shows that correction of the intestinal and dermal dysbiosis, along with appropriate nutritional support and immune modulating therapy, are essential steps in the management of AD.³⁹ However, a Cochrane review failed to find evidence that conventional therapy against S. aureus was clinically helpful in people with AD that is not clinically infected.⁴⁰

Some early studies (for example, Ayers in 1929 and Brown and co-workers in 1935) found a low gastric production of hydrochloric acid was correlated with incidence of AD. Therapy with hydrochloric acid resulted in a dramatic improvement in some cases.⁴¹ A Russian study found markedly reduced activity of membrane-bound small-intestinal enzymes in 346 patients with AD. Correction of this dysfunction resulted in improvements in both digestion and skin.⁴² A related study found a similar problem in infants with AD and reduction of disaccharide intake (e.g. lactose, sucrose and maltose) was instituted.⁴³

A meta-analysis of 21 studies found that current intestinal parasite infection was protective against allergic sensitisation.⁴⁴ This supports the ‘old friends’ hypothesis, a modification of the hygiene hypothesis, where exposure to certain relatively harmless microorganisms (including helminths) supports immunological regulation via gut-associated T regulatory and regulatory dendritic cells.⁴⁵ A less diverse gut microbiota, with high counts of Bacteroides, Clostridium, Enterobacteriaceae and Staphylococcus early in life has been associated with an increased risk for AD.⁴⁶

Based on the above discussion, the following lifestyle and dietary changes can be recommended. Measures to eliminate exposure to house dust mite antigen should be instituted. Simple or multiple exclusion diets should be considered, based on the clinical information. Elimination of cow’s milk (and related dairy products) is recommended as the starting diet if a simple exclusion diet is chosen. Care should be taken to substitute protein and calcium in young children. If there is no symptom improvement in about 4 to 6 weeks, different foods in turn could be tried such as eggs, peanuts or seafood.⁴⁷ Various multiple exclusion diets are available.⁴⁷ These typically involve avoidance of dairy products, eggs, nuts, pork, bacon, shellfish, yeast and fruit. Such diets can be severe and should not be instituted as a first resort. The diet should otherwise be well balanced and should not contain excessive amounts of junk food, sugar and refined carbohydrate.

It is hoped that individual strategies will be devised for each patient with AD, taking some of the insights in this and other chapters into account, but essential elements of herbal treatment are as follows:

Chronic urticaria

Chronic urticaria can be defined as the occurrence of transient wheals lasting more than 6 weeks in duration. In many cases a specific cause is not identified and this is classified as chronic idiopathic urticaria. A physical trigger (heat, cold, exercise, etc.), vasculitis or systemic disease (mainly autoimmunity, especially thyroid disease) can act as a cause.⁴⁹ Associations with Helicobacter pylori, candida infection, malignancy and food intolerances have been reported. An elimination diet focusing on salicylate and amine sensitivities should be considered on a trial basis.

A subset of patients with chronic idiopathic urticaria has autoantibodies to the high affinity IgE receptor, or, rarely, anti-IgE antibodies. Autologous serum skin testing for these autoantibodies can be performed.⁴⁹ These autoantibodies have been shown to activate blood basophils and skin mast cells in vitro. Activation of basophils or mast cells causing histamine release is quite specific for chronic urticaria and defines the autoimmune subgroup.⁵⁰ New research suggests that in some patients the activation of the extrinsic coagulation pathway with thrombin generation might play an important role.⁵¹

It is hoped that individual strategies will be devised for each patient with chronic urticaria, taking some of the insights in this and other chapters into account, but essential elements of herbal treatment are as follows:

Herpes simplex and shingles

Herpes simplex skin outbreaks are characteristic mucocutaneous lesions which are caused by new infection with herpes simplex virus 1 (HSV-1) or 2 (HSV-2), or by reactivation of latent virus residing in the nervous system. Typically HSV-1 affects the face, lips or mouth and HSV-2 affects the genitals, although either virus can infect either location.

HSV is a double-stranded DNA, enveloped virus. HSV infection of some neuronal cells does not result in cell death. Instead, viral genomes are maintained by the cell in a repressed state compatible with survival and normal activities of the cell, a condition known as latency. Immune competency is of primary importance in preventing and dealing with outbreaks caused by latent virus.

The treatment approach for shingles is basically the same as for herpes simplex infection.

Phytotherapy should be centred on the following:

Warts

Warts are a common skin condition, especially before the age of 20. They result from infection with the DNA human papillomavirus, of which there are numerous subtypes. Transmission is by contact with the virus either in live or shed skin.

Most warts eventually resolve spontaneously in healthy people. In immunocompromised patients they can persist and spread (this highlights the importance of immune function). The spontaneous resolution probably results when the immune system ‘discovers’ the virus.

The general approach to treating infections has been provided in Chapter 8. Treatment for warts centres on:

Tinea pedis

Tinea pedis or athlete’s foot is the most common type of human fungal infection. It appears to be spread through sharing of bathing facilities and via swimming pools. Occlusive footwear and infrequent washing of socks can lead to chronicity and relapses. The most common organisms involved include Trichophyton rubrum and T. mentagrophytes.

Phytotherapy is centred on:

If the toe nails are involved, tea tree oil application to the quick of the nail (eponychium) is particularly useful, but in chronic cases may need to be applied for up to 3 months.

Acne

Acne is the most common disease of the skin.⁵³ It affects 85% of teenagers, 42.5% of men, and 50.9% of women between the ages of 20 and 30 years. The role of hormones, particularly as a trigger of sebum production and sebaceous growth and differentiation, is well known. Excess production of hormones, specifically androgens, growth hormone, IGF-1 (insulin-like growth factor 1), insulin, CRH, and glucocorticoids, is associated with increased rates of acne development. Acne may be a feature in many endocrine disorders, including polycystic ovary disease, Cushing syndrome, androgen-secreting tumours and acromegaly. Other non-endocrine diseases associated with acne include Apert syndrome, SAPHO syndrome, Behçet syndrome and PAPA syndrome. Acne medicamentosa is the development of acne vulgaris or an acneiform eruption with the use of certain medications. These medications include testosterone, progesterone, steroids, lithium, phenytoin, isoniazid, vitamins B2, B6 and B12, halogens and epidermal growth factor inhibitors. Management of acne medicamentosa includes standard acne therapy. Discontinuation of the offending drug may be necessary in recalcitrant cases.

Basic medical interventions for acne include topical therapy, systemic antibiotics, hormonal agents, isotretinoin and physical treatments. Generally, the severity of acne lesions determines the type of acne regimen necessary. The emergence of drug-resistant Propionibacterium acnes and adverse side effects are current limitations to effective acne management.⁵³

Acne is a disorder of the sebaceous follicles (sebaceous glands associated with hair follicles), which are located on the face, chest and back. Several factors play a pivotal role in the pathogenesis of acne:

Sebum is the fatty secretion produced by sebaceous glands. Patients with acne have higher rates of sebum production than unaffected individuals and severity of acne can be correlated to sebum production.⁵⁴ Enlargement of the sebaceous glands and increased production of sebum is stimulated by the increase in adrenal and gonadal androgens that precedes the onset of puberty. The first signs of acne commonly occur at this time.

High (but within normal range) serum concentrations of the testosterone precursor dehydroepiandrosterone sulphate (DHEAS) correlate with the presence of acne in prepubertal children and in some adults.⁵⁵ In particular, some women with acne do have raised serum concentrations of testosterone and DHEAS, but in men the situation is less consistent.

Increased local metabolism of DHEAS in sebaceous glands to more potent androgens such as dihydrotestosterone may take place under the influence of androgen-metabolising enzymes such as type I 5alpha-reductase.⁵⁵ Low levels of SHBG in the plasma may act in concert with this process.⁵⁶ This, in addition to a possible role of growth factors, may influence sebum production and hence acne.⁵⁵

Desquamated cornified cells of the upper canal of the sebaceous follicle become abnormally adherent. Instead of undergoing the normal process of shedding, they instead form a microscopic plug (the microcomedo) in the follicular canal.⁵⁵ The mechanism of this process, known as comedogenesis, is not known, but a local deficiency of linoleic acid may be involved.⁵⁷ These lower levels of linoleic acid in sebum may result from accelerated sebum production.⁵⁷

Progressive enlargement of microcomedones gives rise to clinically visible comedones, the non-inflammatory lesions of acne. The follicular canal becomes blocked, but sebum production continues with the resultant proliferation of P. acnes, a normal resident of the sebaceous follicle.⁵⁶ Sebaceous follicles containing microcomedones provide an anaerobic, lipid-rich environment in which these bacteria flourish. Glycerol is a nutritional requirement of P. acnes and is obtained through lipolysis of triglycerides in the sebum, with the release of free fatty acids as a byproduct.

The free fatty acids have proinflammatory and comedogenic properties. P. acnes also releases chemotactic factors, which attract neutrophils to the follicular lumen, resulting in further inflammatory damage.⁵⁸P. acnes further contributes to the inflammatory process through complement activation and release of other hydrolytic enzymes such as proteases. Follicular rupture and extension of the inflammatory process into the surrounding skin results in the formation of the inflammatory lesions of acne – papules, pustules and nodules. Abnormalities in neutrophil chemotaxis and phagocytosis of P. acnes and activation of macrophages contribute to the chronic, haphazard nature of inflammation in severe acne.^59,60

Interest in sebaceous gland physiology and its diseases is rapidly increasing. Exploration of sebaceous gland biology, hormonal factors, hyperkeratinisation, role of bacteria, sebum, nutrition, cytokines and TLRs is receiving research attention. Sebaceous glands play an important role as active participants in the innate immunity of the skin. They produce neuropeptides, excrete antimicrobial peptides and exhibit characteristics of stem cells. Androgens affect sebocytes and infundibular keratinocytes in a complex manner influencing cellular differentiation, proliferation, lipogenesis and comedogenesis. Retention hyperkeratosis in closed comedones and inflammatory papules is attributable to a disorder of terminal keratinocyte differentiation. P. acnes, by acting on TLR-2, may stimulate the secretion of cytokines, such as IL-6 and IL-8 by follicular keratinocytes and IL-8 and IL-12 in macrophages, giving rise to inflammation. Certain P. acnes strains may induce an immunological reaction by stimulating the production of sebocyte and keratinocyte antimicrobial peptides, which play an important role in the innate immunity of the follicle. Qualitative changes of sebum lipids induce alteration of keratinocyte differentiation and induce IL-1 secretion, contributing to the development of follicular hyperkeratosis. High glycaemic load food and milk may induce increased tissue levels of 5alpha-dihydrotestosterone. These new aspects of acne pathogenesis lead to the considerations of possible customised therapeutic regimens.⁶¹

One interesting hypothesis offers a solution for the pathogenesis of acne and explains all major pathogenic factors at the genomic level: this theory is a relative deficiency of the nuclear transcription factor forkhead box-containing transcription factor (Fox)O1.⁶² This poorly understood transcription factor modulates energy homeostasis at both the cellular and whole-body levels. Nuclear FoxO1 suppresses androgen receptor, other important nuclear receptors and key genes of cell proliferation, lipid biosynthesis and inflammatory cytokines. Elevated growth factors during puberty and persistent growth factor signals due to Western lifestyle stimulate the export of FoxO1 out of the nucleus into the cytoplasm via activation of the phosphoinositide-3-kinase (PI3K)/Akt pathway. By this mechanism, genes and nuclear receptors involved in acne are derepressed, leading to increased androgen receptor-mediated signal transduction, increased cell proliferation of androgen-dependent cells, induction of sebaceous lipogenesis and upregulation of TLR-2-dependent inflammatory cytokines. All known acne-inducing factors exert their action by reduction of nuclear FoxO1 levels. In contrast, retinoids, antibiotics and dietary intervention can increase the nuclear content of FoxO1, thereby normalising increased transcription of genes involved in acne.

As touched on above, despite the popular conventional view that diet is not related to acne, several experts in the field are encouraging a reassessment of this position, partly because the data refuting any association are weak.⁶³ For example, one review pointed out that acne is virtually unknown in traditional hunter-gatherers.⁶⁴ The review goes on to suggest an association between dairy products (from the protein, not the fat) and dietary fat (especially in terms of increased sebum production). The severe methodological shortcomings of the negative trials involving chocolate were stressed. Such trials have been used to refute any connection between diet and acne.

Recent evidence has demonstrated that the hormonal cascade triggered by diet-induced hyperinsulinaemia elicits an endocrine response that simultaneously promotes unregulated tissue growth and enhances androgen synthesis.⁶⁵ This former aspect is regulated by IGF-1, a potent mitogen, required for keratinocyte proliferation. Another review considered that there was compelling evidence that high glycaemic load diets may exacerbate acne.⁶⁶

The view is growing that acne can be regarded as an indicator disease of the exaggerated insulinotropic effects of modern Western diets.⁶⁷ Especially milk- and whey protein-based products contribute to elevations of postprandial insulin and basal IGF-1 plasma levels. This is a key aspect of mammalian milk designed to promote growth in the neonate. Increased insulin/IGF-1 signalling reduces nuclear FoxO1.⁶⁷ IGF-1 stimulates 5alpha-reductase, adrenal and gonadal androgen synthesis, androgen receptor signal transduction, sebocyte proliferation and lipogenesis.⁶⁸

The use of strong soaps and shampoos that strip sebum from the skin and hair and initiate a rebound increase in sebum production should be avoided. Herbal extracts such as licorice, which are rich in saponins, can act as a gentle skin wash which will not cause a rebound in sebum production.

Appropriate dietary strategies for acne, as supported by the above evidence, stress the avoidance of high glycaemic index foods and dairy products. The diet should also not contain excessive levels of fat.

It is hoped that individual strategies will be devised for each patient with acne, taking some of the insights in this and other chapters into account, but essential elements of herbal treatment are as follows:

Assisting healing and repair

The basic response of all living organisms to damage or loss of body tissues is to initiate repair and regenerate the damaged part. So while surgeons may claim that they save the lives of their patients (and they often do), without this capacity of the body to heal itself surgery would in fact be a death sentence.

There are three basic processes involved in wound healing:⁷⁴

When part of the body is wounded the essential sequence of events is as follows. The damaged tissue and the bleeding around it stimulate the body’s inflammatory response. White blood cells (particularly macrophages and neutrophils) invade the damaged tissue in large numbers and begin engulfing (by phagocytosis) and processing the large amounts of debris. This usually takes several days. During this process fine and delicate new blood vessels begin to form by the process known as angiogenesis. These new blood vessels provide nutrition to support the next stage of healing, which is the regeneration of connective tissue. During this time other necessary tissues such as nerve cells and lymph vessels begin to form.

Fibroblasts are the cells that synthesise the non-cellular constituents of connective tissue, namely collagen and glycosaminoglycans, the jelly-like ground substance into which the collagen is embedded.

With the passage of time there is organisation and remodelling of the new connective tissue, which becomes progressively denser. The zone of new dense connective tissue is known as a scar. The fibroblasts contract, pulling the margins of the wound together. Scar remodelling continues for months or years after the wound has healed,⁷⁴ and this suggests a long-term role for herbs such as gotu kola to minimise the scar.

Regeneration of connective tissue is half of wound healing; the other half is restoration of the cells that line the surfaces of tissue (known as epithelium). Skin cells are a form of epithelium and the skin cells around the margin of a wound begin to multiply and cover the gap created by the wound.

Other tissues (apart from skin) regenerate by a similar process (if they are capable of doing so). For example, bone heals in a very similar way, except that osteoblasts are involved. Hence the approach suggested below should also assist bones to heal.

If the scavenging phase of the macrophages is disturbed, say by infection, healing will be slow. Also, a good blood supply to the general area is essential for efficient healing. Excess movement of the wound edges can also impair healing; this is the main reason why broken limbs are placed in a cast.

Hence the key steps in healing are:

The following herbs can assist these steps: