Hypertrophic osteoarthropathy (HOA) is a syndrome characterized by abnormal proliferation of the skin and osseous tissues at the distal parts of the extremities. Three clinical features are typically present: a peculiar bulbous deformity of the tips of the digits conventionally described as 'clubbing', periostosis of the tubular bones and synovial effusions.
Drumstick fingers are so unique that its recognition usually poses no dilemma. Diagnostic criteria for HOA are the combined presence of clubbing and radiographic evidence of periostosis of the tubular bones. Synovial effusion is not essential for the diagnosis. Nevertheless, it should be emphasized that in some patients - particularly those with malignant lung tumors - painful arthropathy may be the presenting manifestation of the syndrome, in advance of clubbing. Such cases could be misdiagnosed as suffering from an inflammatory type of arthritis. Here, important clinical features in the differential diagnosis are the location of pain - in HOA not only the joint is involved, but also the adjacent bone - plus the fact that rheumatoid factor is usually absent and synovial fluid is 'non-inflammatory' in nature.
THREE DECADES STUDYING THE OLDEST CLINICAL SIGN IN MEDICINE,
National Institute of Cardiology. Mexico
Hippocrates (island of Cos, Greece circa 460 BC) is known as the founder of Medicine. He based his medical practice on observations and on the study of the human body. Hippocrates held the belief that illnesses were not a divine contrive but had a physical and a rational explanation thus laying the foundations of scientific Medicine. He was the first to document digital clubbing as a sign of disease. He described: “water accumulates; the patient has fever and cough; the respiration is fast; the feet become edematous; the nails appear curved and the patient suffers as if he had pus inside...if you put your ear against the chest you can hear it seethe inside like sour wine”. This is the reason why digital clubbing is also known by the eponym “Hippocratic finger” and is regarded as to be the oldest clinical sign in Medicine (1).
Through the centuries digital clubbing has been known to be an ominous sign, heralding the presence of a dreadful internal illness. With the introduction of x rays imaging in the practice of Medicine in the late nineteenth century, it was discovered that this unique finger deformity is often accompanied by periosteal proliferation of the tubular bones. Bamberger (2) and Marie (3) named the fully developed expression as pulmonary hypertrophic osteoarthropathy. The suffix “pulmonary” was latter abandoned when it was realized that the skeletal syndrome may also appear as a response of severe diseases located outside the lungs and even may be present in otherwise healthy individuals (“primary hypertrophic osteoarthropathy”). Digital clubbing and hypertrophic osteoarthropathy (HOA) represent different stages of the same syndrome (4).
Digital clubbing is a clear-cut unambiguous clinical sign easily recognized by medical students and physicians alike. In contrast the mechanisms whereby so different severe internal illnesses can induce this unique finger deformity have remained unidentified.
This article narrates the chronology of our research on the pathogenesis of clubbing. Emphasis is given to the fact that careful study of unique clinical cases provided important clues that were the basis for fruitful prospective controlled investigations.
Our advances in the study of clubbing/HOA:
During In the late seventies, soon after returning from a five-year period training in internal medicine and rheumatology in the United States of America, I started working at the National Institute of Cardiology of Mexico. There was a youngster with congenital heart disease who had diffuse swelling of the legs not ascribable to heart failure. She was deeply cyanotic with marked clubbing of the twenty digits. X rays demonstrated the cause of the leg swelling was a conspicuous thickening of the tibial bones (5). At that time it was held true that HOA was rare in cases of cyanotic heart diseases. This was the main argument used in favor of the notion that clubbing and HOA were different syndromes. A prospective investigation demonstrated that all patients with cyanotic heart diseases have digital clubbing and over a third of them (those with the severest extrapulmonary shunting of blood) develop full blown HOA (6). No other internal illness is so highly associated to HOA.
At that time (late seventies) the pathogenic theories to explain the acropachy were based on anecdotes. For instance, it was proposed that HOA developed from an impulse originated in the diseased organ travelling the vagus nerve, and that vagotomy cured HOA. This theory (still mentioned nowadays in some textbooks) is based on a single report of 5 cases with unresectable lung cancer. In these cases, joint pain and swelling improved after cutting the vagus nerve during thoracotomy, but there was no not claim of clubbing regression (7). Similar symptomatic improvement has been achieved by laparotomy without severing the vagus nerve (8).
In 1987 We proposed a unifying hypothesis (9) stating that clubbing/HOA appears appear when the lung fails to inactivate a fibroblast growth factor normally present in the venous circulation. The hypothesis was based on two pillars: 1) the existence of patients with patent ductus arteriosus with reversal of flow, in whom clubbing is limited to the cyanotic limbs. 2) The discovery at that time of the biology of different fibroblast growth factors that could theoretically explain the histological alterations of clubbing/HOA. For the lung cancer cases it was proposed that a tumor-derived growth factor would gain direct access to the systemic circulation thus inducing clubbing. Since diverse internal illnesses induce a unique finger deformity, it was also suggested that different pathogenic mechanisms generate a single clubbing promoting growth factor.
Later on, Dickinson and Martin anticipated that such growth factor could be platelet-derived (10). They based their theory on an iconoclastic mathematical model that proposed that in normal circumstances all platelets are formed in the highly subdivided lung vasculature as result of progressive megakaryocyte fragmentation. This is the mathematical explanation for the distinctive log-normal configuration of the platelet volume distribution curves. In cases with extrapulmonary shunting of blood, large platelet fragments would gain direct access to the systemic circulation, reaching its most distal sites on axial streams there activating endothelial cells releasing growth factors and inducing clubbing. Our studies in patients with cyanotic heart disease concur with this explanation. Such patients have macrothrombocytes in their peripheral circulation with distorted volume-distribution curves (11). There is now additional, more definitive, evidence of the pulmonary origin of platelets (12). The finding of high circulating levels of Von Willebrand factor antigen in patients with HOA reinforces the notion of enhanced platelet/endothelial cell activation (13). Morphological studies proved that HOA was an orderly process that started at the most distal sites of the extremities with the appearance of digital clubbing and evolving in a centripetal fashion with thickening of the tubular bones (14,15).
The nature of the purported growth factor responsible for clubbing remained to be established. We tested platelet-derived growth factor in carefully separated plasma from patients with different types of HOA, with negative results (16). Once again the study of a fascinating case provided us with an important clue. A 64 year/old gentleman suffered from severe peripheral neuropathy. Physical examination demonstrated digital clubbing (he was unaware of the presence of the deformity). Laboratory tests unveiled a monoclonal gammopathy. The diagnosis was an “incomplete” POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, M component and skin changes). There was little information on the presence of clubbing in POEMS syndrome. So, the results of a metaanalysis searching for this association were surprising (17). Clubbing was present in one third of all patients with POEMS syndrome reported in the literature. This proportion is likely to be an underestimation, because in some instances the finger deformity could be overlooked or dismissed, and thus not being included in the report. Additionally this metaanalysis found that patients with POEMS syndrome frequently display other HOA features such as pachyderma and hyperhidrosis. There is clear overlap between HOA and POEMS syndrome.
At the time of the metaanalysis (1997), two independent groups of investigators described that patients with POEMS syndrome have very high circulating levels of a recently described growth factor; vascular endothelial growth factor (VEGF) (18,19). The biologic properties of this cytokine fitted correctly into the pathogenesis of HOA. VEGF is a platelet derived growth factor, its action is induced by hypoxia, and it is produced by diverse malignant tumors fostering their uncontrolled growth. On the other hand VEGF promotes edema, vascular hyperplasia, fibroblast proliferation and new bone formation (20, 21); such are HOA specific histological changes (4).
We undertook a prospective investigation to determine the plasma VEGF levels in three different groups of patients with HOA; those associated to lung cancer, to cyanotic heart disease and cases with primary HOA. These patients were compared to three controls groups: healthy individuals, patients with lung cancer without HOA, and patients with emphysema without HOA. In all instances plasma was separated with the Schiebout-Clark method in order to avoid artificial platelet activation. Results showed two variables associated to significantly raised plasma levels of VEGF: the presence of HOA and the presence of cyanosis. Patients with lung cancer that displayed HOA and those with primary HOA had significant increased VEGF levels in their plasma when compared to healthy controls. (22).
Another intriguing clinical case reinforced the notion that VEGF plays a central role in the development of clubbing/HOA. The case of a 44 year/old female that complained of severe bone pain of several months duration. Physical examination demonstrated marked digital clubbing. Bone radiographs showed periosteal proliferation of tibias and fibulae. Chest radiographs showed a poorly define defined left apical mass. She had extremely high VEGF plasma levels (390 pg./ml) (normal range 0-26 pg./ml). She underwent a wide surgical excision of the lung mass. Histology showed non-small cell adenocarcinoma. After surgery there was a dramatic disappearance of all HOA features including digital clubbing and her postoperative plasma VEGF fell to normal levels. Histochemistry study of the excised tumor confirmed abnormally high VEGF expression (23).
More recently Atkinson and Fox gave additional credibility to the notion that VEGF plays a key role in the development of the acropachy. Post-mortem immunohistochemistry demonstrated that VEGF and its receptor were overexpressed in stromal fibroblast of five clubbed specimens but not in controls (24).
The proposal of VEGF in the epicenter of clubbing/HOA could explain how different hypoxic or malignant diseases could induce the acropachy. As already stated VEGF is a platelet-derived factor induced by hypoxia. Also it is abnormally produced by diverse malignant tumors fostering their uncontrolled growth (20,21).The multiple diseases associated to clubbing/HOA have in common the presence of prominent platelet and/or endothelial cell activation. In practically all of them enhanced VEGF expression have has been reported. This is true for cyanotic heart diseases, diffuse interstitial lung pathologies, Graves’ disease, mesothelioma, liver cirrhosis and lung cancer among many others. On the other hand the distinguishing biological effects of VEGF provide a coherent explanation for the peculiar histological alterations of clubbing/HOA (vascular hyperplasia, edema and excessive fibroblast and osteoblast formation). VEGF has demonstrated to be a potent osteogenic agent. A stronger argument for the principal role of VEGF in clubbing development comes from the positive results of recent studies looking specifically at this issue. Patients with digital clubbing of different origin, have high VEGF plasma levels (22,23) and enhanced VEGF tissue expression (24).
Emerging evidence proposes that VEGF plays a central in the pathogenesis of clubbing/HOA. The discovery of the etiology of the oldest clinical sign of Medicine is not only of archaeological interest. It would surely help to understand the pathogenesis of multiple underlying illnesses that bring on such peculiar and unique finger deformity.
www.martinez-lavin.com |Todos los derechos reservados