The role of microvascular disease in diabetic foot ulceration – Clinical
Diabetic foot ulcers are a major cause of morbidity in diabetes. This article reviews the literature assessing the role of microvascular disease in the development of diabetic foot wounds. Several elements are considered, among them, basement membrane thickening, reduction in maximal hyperaemia, microvascular pressure, external pressure and increased blood flow. Overall, there is little evidence to implicate microvascular changes alone as sufficient to result in capillary closure.
Foot wounds represent the most important of all the long-term problems of diabetes in medical, social and economic terms. The risk of developing foot ulceration in diabetes is greater than reaching the end-stage sequelae of retinopathy or nephropathy (Boulton et al, 2000). The aetiology of diabetic foot lesions is a complex issue, but it is recognised that three key elements are involved: neuropathy, peripheral vascular disease and infection (Kamal et al, 1996). An intact and functional microvascular system is essential for tissue nutrition, inflammatory responses and removal of waste products, but although microvascular disease is accepted as playing a part in diabetic foot ulceration, evidence that it is a major causative factor remains contentious.
Key role for microvascular disease?
Goldenburg et al (1959) suggested that the main cause of diabetic foot ulceration was arteriolar occlusive disease leading to ischaemic lesions in the presence of normal pedal pulses.
Goldenburg et al’s (1959) theory has since been disputed by a number of subsequent investigators who were unable to find a difference in the amount of microvascular disease between diabetic and non-diabetic amputees (Conrad, 1967; LoGerfo and Coffman, 1984).
Strandness et al (1964) carried out a similar study to Goldenburg et al (1959) on amputated limbs, but were also unable to substantiate their theory.
Crane and Branch (1998) stated that microvascular disease does not exist and that peripheral vascular disease is atherosclerotic in nature affecting the more proximal vessels while sparing the distal vessels. This sentiment was supported by Banga (1995), who suggested that there was no convincing histological or physiological evidence of occlusive microvascular disease in the limbs of diabetics, and that studies of the microcirculation have only shown that diabetes affects perfusion at different stages of lower limb ischaemia.
On the other hand, easily palpable foot pulses and the absence of occlusive arterial diseases in the neuropathic foot directly implicate the microcirculation as a causative factor in the development of diabetic foot ulceration (Flynn and Tooke, 1992). Impairment of the hyperaemic response has been shown to be closely associated with retinopathy rather than neuropathy or duration of diabetes (Walmsley et al, 1989), and such responses suggest a mechanism through which microvascular disease predisposes to foot ulceration.
A case report by Tesfaye and Gill (1997) concluded that there is evidence of small vessel disease in the aetiology of diabetic foot ulceration in the absence of macroangiopathy. This study looked at only two patients. Although the results suggested small vessel disease alone was responsible for the patient’s ulceration, the authors did not address the possibility of vessel calcification, which may have falsely implied strong foot pulses.
Basement membrane thickening
Although its aetiology is not fully understood, a thickening of the basement membrane is a consistent finding in the capillaries (LoGerfo and Coffman, 1984), and arterioles (Silhi, 1998) of diabetic patients. It also tends to increase with duration of diabetes, age and the distance of the vessel below the heart (Williamson et al, 1998). Thickening of the basement membrane is thought to be a result of the proliferative response of the vascular endothelium to chronic injury (Silhi, 1998), brought on by the effects of hyperglycaemia on glucose metabolism, protein synthesis and non-enzymatic glycosylation of cellular and extracellular components.
The accumulation of sorbitol in the cells due to defective glucose metabolism (Frank, 1994) leads to increased oedema (Watkins et al, 1997), resulting in modified membrane function and permeability. Alterations in basement membrane protein synthesis and advanced glycation end products (AGEs) formed as a result of nonenzymatic glycosylation lead to functional changes of the microcirculation, causing a reduced ionic charge, allowing albumen across the basement membrane and leading to altered cellular nutrition (Stonebridge, 1996).
Increased oxidative stress increases collagen cross linking, rendering the membrane less soluble and less susceptible to proteolytic turnover (Feener et al, 1997).
Ferguson et al (1996) carried out a relatively minor study examining biopsies taken from the margins of diabetic foot ulcers and limb amputations. They noted that although the majority of wounds were well vascularised, there was narrowing or occlusion of some large vessels due to membrane wall thickening in over half of their subjects. They also observed the appearance of von Willebrand factor in the interstitial spaces suggesting an alteration in the functioning of these vessels in diabetic patients. A cell culture plate does not, however, represent the intravascular environment–in which cellular elements come into contact with the vessel wall, and dynamic phasic changes in pressure, flow, stretch and shear occur and interactions with subjacent tissues take place.
Timperley et al (1985) compared II diabetic patients with peripheral neuropathy but good glycaemic control with six subjects with non-diabetes related peripheral neuropathy, and a further control group comprising 6 subjects with no evidence of peripheral nerve disease. Electron microscopical examination showed hyperplasia of endothelial cells in some small endoneural blood vessels in all the diabetic patients examined. Plugging of the lumen by the cells was observed in seven cases; in one case, the lumen of a vessel was occluded by thrombus, and fibrin was seen to be tracking through the vessel wall. In another vessel, a plug of degranulated platelets was seen; in another vessel, red blood cells were seen in the vessel wall. In controls, vessels were normal except for one man with carcinomatous neuromyopathy. Electrophsiological studies showed a pattern of denervation suggesting a relationship between neuropathy and multiple small vessel infarcts. The study demonstrates that small vessel abnormalities play a n important role in the damage of other tissues in diabetes.
However, caution should be exercised when transferring results from studies carried out on different locations in the body (often, endothelial cells from arteries and veins rather than microvasculature are used, and cell passage modifies cellular response to a variety of stimuli).
Reduced maximal hyperaemia
A reduction in maximal hyperaemia has also been observed in diabetic patients, and it has been hypothesised (Wall, 1997) that this is due to the fact that there are fewer capillaries, a phenomenon that occurs in glomerulonephropathy, although this has not been observed with clinical video microscopy. Histological evidence, however, confirms that impaired hyperaemia is associated with basement membrane thickening (Tooke and Brash, 1995), although further investigation into the mechanisms mediating this response are required.
Shore et al (1991) investigated microvascular disease in the absence of macrovascular disease, comparing 50 diabetic children with 50 non-diabetic children. Laser Doppler fluximetry was used to measure the maximum hyperaemic response to direct local heating of 44[degrees]C on foot skin. Their study revealed an impaired hyperaemic response in diabetic children, which was directly related to duration of diabetes, but not to long-term blood glucose control, suggesting that functional abnormalities of the microcirculation can occur independently of large vessel disease. This study also highlighted that the prevalence of diabetic complications was greater after puberty. However, it is not clear whether pre-pubescent children are protected from microvascular disease or whether the clinical criteria used is unable to detect subtle microvascular changes.
Rayman et al (1986), confirming Shore’s (1991) theory, found an impaired hyperaemic response to two types of stimuli in type I diabetic patients with no evidence of microvascular disease. This small study on older subjects compared 23 type I diabetic patients (mean age of 30.9 years) against 21 healthy control subjects (mean age of 29.7 years). To determine whether capillary density was related to differences in flow, laser Doppler flowmetry was used to determine microvascular perfusion, while television microscopy was used to calculate the number of superficial capillary loops per [mm.sup.2]. Although peak blood flow did not correlate with age in either group, it was inversely related to the duration of diabetes. The mean density of capillaries was similar in both groups and capillary numbers did not correlate with blood flow. The authors postulated that the impaired vasodilator response was a failure in the release of local vasoactive mediators, a theory reinforced by the finding that substance P, a potent neurogenic vasodilator, is depleted in diabetic nerves. Another explanation could lie in the thickened basement membrane and reduced dilation, characteristic of long-term diabetes, limiting the microcirculation to react to the vasoactive mediators.
Such studies are less reliable in type 2 diabetic patients as there is more likelihood of large vessel disease due to its pathogenesis along with other medical complications associated with age and lifestyle, which may have preceded the diagnosis of diabetes. Atherosclerosis and hypertension are far more prevalent in non-insulin dependant diabetics and it has been hypothesised (Tooke and Brash, 1995) that this is due to prolonged periods of hyperinsulinaemia before diagnosis.
Fagrell et al (1984) used vital capillaroscopy to study the flow velocity in the microcirculation of 14 diabetic patients and 14 non-diabetic matched controls. The results confirmed that capillary blood flow was equal in both groups, but a delayed hyperaemic response was evident in the diabetic subjects. This may have been due to a vasomotor dysfunction in the precapillary arterioles or an increased blood viscosity, but its occurrence correlated with retinopathy, nephropathy and neuropathy in the insulin-controlled diabetic subjects.
Capillary pressure is determined by the ratio of pre- to post-capillary resistance, and changes in microvascular pressure and flow do not necessarily parallel each each another (Tooke, 1996). The link between reduced blood flow in nutritive capillaries and diabetic complications, such as retinopathy and nephropathy, has been suggested as the true microangiopathy (Rendell and Bamisedun, 1992).
Rendell and Bamisedun (1992) demonstrated that reverse flow capacity in the nutritive capillaries of diabetic patients is reduced at high temperatures, and is correlated with duration of diabetes and other microvascular complications, but not with neuropathy. The greatest severity of this was demonstrated in the lower extremities. This would suggest that diabetic cutaneous microangiopathy coexists with diabetic renal and retinal microvascular disease. The authors proposed that this was directly related to increased pressure in the microvasculature.
A decreased blood flow observed in the toe pulp of the diabetic subjects with no evidence of neuropathy is more difficult to explain, but is possibly due to decreased red blood cell velocity or to abnormality of the red blood cells themselves. The age of the participants potentially rules out arteriosclerosis as an obvious cause.
Tooke (1986) suggests that the more rigid diabetic erythrocyte could exert injurious tangential pressure on the vessel wall affecting microvascular blood flow. It was demonstrated (Sandeman et al, 1992) that nail fold capillary pressure is elevated in insulin-dependant diabetic patients from an early stage and this is positively related to glycaemic control compared with patients not demonstrating diabetic complications.
Effect of pressure
The following studies were carried out on the effects of pressure on underlying muscle degeneration.
Husain (1953) highlighted microscopic evidence of cellular infiltration, interstitial capillary haemarrhage and various stages of cellular degeneration in the muscles of rats, when pressure of 100 mmHg was applied for one hour, emphasising the ease of which local ischaemia may occur.
Kosiak (1959) also found that the higher the pressure, the more rapidly the skin ulcerated, when applying pressures of 100-550 mmHg to the skin of dogs for periods of 1-2 hours.
These early studies demonstrated that microscopic degenerative changes occur when tissue is subject to pressure for short periods of time, but that the normal hyperaemic response more than compensates for the temporary ischaemia, and relief of the pressure may permit restoration of normal cellular metabolism without ulceration, in healthy tissue.
Thus, it is the duration of pressure more than the intensity that leads to long-term damage. Although these experiments were performed in a controlled clinical environment on healthy animals, they demonstrated the damaging effects of pressure on the microcirculation.
The foot by its very nature is routinely exposed to different pressures including shear and friction forces and although it is possible that pressure ischaemia is preceded by neuropathy, it undoubtedly plays an important role in the development of foot ulceration. In people with diabetes, increased pressure due to ill-fitting footwear or deformity on an insensate foot may allow increasing periods of strain leading to tissue breakdown. It has been documented that pressure relief is the first line approach in the treatment of diabetic foot ulceration (Tyrrell 1999) leading to a good prognosis.
Capillary closure may be due to extrinsic pressure or precapillary vasospasm, but there is no histological evidence of empty skin capillaries in the diabetic foot (Flynn and Tooke, 1992) from studies carried out on eyes or skeletal muscles. Although studies of other organs are useful, their relevance to the diabetic foot is questionable since microvascular beds are different, especially in their control mechanisms from organ to organ (Tooke, 1996).
A study by Moriarty et al (1994) assessed capillary circulation in the feet of 23 patients with diabetic neuro-ischaemic ulcerations. The method of 99mTcmacroaggregated albumin perfusion scanning was used, which allows the advantage of showing the blood flow to all the capillaries of the foot. It was shown that increased perfusion around the ulcer was significantly associated with successful healing, whereas poor capillary perfusion around the ulcer was associated with poor healing. The study also suggested that adequate perfusion to the deeper tissues, such as bone and muscle beneath the ulcer, was equally important as adequate perfusion to the edges of the ulcer to determine successful healing.
These studies do not confirm whether external pressure or vasospasm is a cause of microvascular disease in the foot, and further research is required to determine whether this is the case. It does, however, demonstrate that insufficient microvascular perfusion can delay or inhibit healing.
Tooke (1986) proposes that rather than a reduction in blood flow, an increase in blood flow through the capillaries is observed in early diabetes, generating a range of vasoactive mediators by the endothelium. Failure of pressure regulation and reduced maximum hyperaemia may explain increased blood flow at rest, especially in the dependent position that is a feature of limb ischaemia. It is also thought to lead to the production of nitric oxide (Silhi, 1998) as a result of the vessel wall tension, leading to endothelial cell damage and eventual basement membrane thickening. This theory is supported by the Diabetes Control and Complications Trial (DCCT) (1993), which reported an increase in levels of fibrinogen and total globulin in patients with diabetic retinopathy and nephropathy, predisposing to red cell aggregation.
Long-term hyperglycaemia is recognised as being responsible for many diabetes-related defects. Reichard et al (1993) studied the relationship between blood glucose concentrations and microvascular complications in a random selection of 102 insulin-dependent diabetic patients. The patients were split into two groups, one receiving intensive insulin treatment and the other, standard insulin treatment. All participants were considered to have unsatisfactory blood glucose control at the start of the trial. The results highlighted a lag between high blood glucose concentrations and the appearance of serious retinopathy.
After 94 months, serious retinopathy was more common in the standard treatment group, and its appearance was related to the glycosylated haemoglobin values during the first five years of the study. This demonstrated that lower blood glucose concentrations retarded the development of microvascular complications even when retinopathy had begun to develop.
The development of nephropathy and neuropathy were also shown to be retarded after a few years of intense insulin treatment. These results suggest a direct link between hyperglycaemia and the aetiology of microvascular disease.
Damage of sympathetic nerve fibres is recognised early in diabetes due to peripheral neuropathy. It is well documented (e.g. Tooke, 1995) that this results in arteriovenous shunting that effectively bypasses the capillaries. However, there is no supporting evidence to suggest that this affects the microcirculation.
Patients with neuropathy tend to have warm flushed feet due to venous congestion and blood passing through the arterio-venous shunts. Despite this apparent increased blood flow, the tissues tend to remain anoxic due to the blood bypassing the capillaries and hence giving up little of its oxygen (Bliss, 1998). Impaired pressure regulation causes an increase in oedema, which will consequently increase intracapillary diffusion distances, and increase interstitial fluid pressure, which may compress the capillaries. It is also likely that external pressure from footwear will also damage the capillaries if the patient does not modify their footwear to accommodate the oedema.
Metabolic and haemodynamic factors both play a role in the aetiology of diabetic microangiopathy and may indeed be synergistic. Despite reservations, sufficient evidence exists to implicate the microvascular endothelium as responsible for a variety of pathophysiological phenomena observed in diabetes, but whether it is the underlying element accountable for the development of diabetic foot ulceration has yet to be determined.
There is little evidence to implicate microvascular changes alone as sufficient to result in capillary closure. Thus, failure of the microcirculation is possibly more likely to be linked with increasing duration of diabetes and functional abnormalities of the microvasculature developing in parallel with neuropathy and occlusive arterial disease. The inability to generate a hyperaemic response in diabetes results in increasing underperfusion during times of tissue stress, which must ultimately compromise the body’s healing potential.
Although microvascular disease evidently has a major impact on tissue breakdown, there is as yet no firm evidence that it plays a primary role in the development of diabetic foot ulceration.
I would like to acknowledge the Wound Healing Research Unit at the University of Wales College of Medicine, where I am studying for my MSc degree, and the Diabetes Team with whom I work at North Bristol NHS Trust.
Banga JD (1995) Lower extremity arterial disease in diabetes mellitus. Journal of British Podiatric Medicine 50(5): 68-72
Bliss MR (1998) Hyperaemia. Journal of Tissue Viability 8(4): 4-12
Boulton AJM, Connor H, Cavanagh PR (2000) The Foot in Diabetes, 3rd edn. John Wiley and Sons Ltd. Chichester
Conrad MC (1967) Large and small artery occlusion in diabetics and nondiabetics with severe vascular disease. Circulation 36: 83-91
Crane M, Branch P (1998) The healed diabetic foot what next? In Harkless LB, Armstrong DG, Fish SE (1998) Healing the diabetic wound. Clinics in Podiatric Medicine and Surgery 15:1: 155-164
Diabetes Control and Complications Trial Research Group (1993) The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. New England Journal of Medicine 329: 977-86
Fagrell B, Hermansson I-L, Karlander S-G, Ostergren J (1984) Vital capillary microscopy for assessment of skin viability and microangiopathy in patients with diabetes mellitus. Acta Med Scand Suppl 687: 25-8
Feener E, King P, George L (1997) Vascular dysfunction in diabetes mellitus. The Lancet 350: (suppl 1) 9-13
Ferguson MWJ, Herrick SE, Spencer MJ, Shaw JE, Boulton AJM, Sloan P (1996) The histology of diabetic foot ulcers. Diabetic Medicine 13: S30-S33
Flynn MD, Tooke JE (1992) Aetiology of diabetic foot ulceration: a role for the microcirculation. Diabetic Medicine 8: 320-9
Frank RN (1994) The aldose reductose controversy. Diabetes 43: 169-72
Goldenburg S, Alex M, Joshi RA, Blumenthal HT (1959) Nonatheromatous peripheral vascular disease of the lower extremity in diabetes mellitus. Diabetes 8(4): 261-73
Husain T (1953) Experimental study of some pressure effects on tissue, with reference to bed sore problems. Journal of Pathology and Bacteriology 66: 347
Kamal K, Powell RJ, Sumpio BE (1996) The pathology of diabetes mellitus: Implications for surgeons. Journal of the American College of Surgeons 183: 271-89
Kosiak M (1959) Etiology and pathology of ischaemic ulcers. Arch Physic Med and Rehab Feb: 62-9
LoGerfo FW, Coffman JD (1984) Vascular and microvascular disease of the foot in diabetes. The New England Journal of Medicine 311(25): 1615-19
Moriarty KT, Perkins AC, Robinson AM, Wastie ML, Tattersall RB (1994) Investigating the capillary circulation of the foot with 99m Tc-macroaggregated albumin: a prospective study in patients with diabetes and foot ulceration. Diabetic Medicine 11: 22-7
Rayman G, Williams SA, Spencer PD, Smaje LH, Wise PH, Tooke JE (1986) Impaired microvascular hyperaemic response to minor skin trauma in type 1 diabetes. British Medical Journal 292: 1295-8
Reichard P, Nilason B-Y, Rosenqvist U (1993) The effect of long-term intensified insulin treatment on the development of microvascular complications of diabetes mellitus. The New England Journal of Medicine 5: 304-309
Rendell M, Bamisedun O (1992) Diabetic cutaneous micro-angiopathy. American Journal of Medicine 93: 611-8
Sandeman DD, Shore AC, Tooke JE (1992) Relation of skin capillary pressure in patients with insulin-dependant diabetes mellitus to complication of metabolic control. New England Journal of Medicine 372: 760-4
Shore AC, Price KJ, Sandeman D, Green EM, Tripp JH, Tooke JE (1991) Impaired hyperaemic response in children with diabetes mellitus. Diabetic Medicine 8: 619-23
Silhi N (1998) Diabetes and wound healing. Journal of Wound Care 7(l): 47-51
Stonebridge PA (1996) Surgical management of peripheral vascular disease in diabetes. Diabetes Reviews International 5(4): 9-12
Strandness DE, Priest RE, Gibbons GE (1964) Combined clinical and pathological study of diabetic and non-diabetic peripheral arterial disease. Diabetes 13: 366-72
Tesfaye S, Gill GV (1997) Small vessel disease: a cause of foot ulceration in the neuropathic foot? Practical Diabetes International 14(3): 78-9
Timperley WR, Boulton AJM, Davies-Jones GAB, Jarratt JA, Ward JD (1985) Small vessel disease in progressive diabetic neuropathy associated with good metabolic control. Journal of Clinical Pathology 38: 1030-8
Tooke JE (1986) Microvascular haemodynamics in diabetes mellitus. Clinical Science 70: 119-25
Tooke JE, Brash PD (1995) Microvascular aspects of diabetic foot disease. Diabetic Medicine 13: S26-S29
Tooke JE (1996) Peripheral microvascular disease in diabetes. Diabetes Research and Clinical Practice 30 (suppl): S61-S65
Tyrrell W (1999) Orthotic intervention in patients with diabetic foot ulceration. Journal of Wound Care 8(10): 530-2
Wall B (1997) Assessment of ischaemic feet in diabetes. Journal of Wound Care 6(1): 32-8
Walmsley D, Wales JK, Wiles PG (1989) Reduced hyperaemia following skin trauma: evidence for an impaired microvascular response to injury in the diabetic foot. Diabetologia 32: 736-9
Watkins PJ, Drury PL, Howells SL (1997) Diabetes and its Management, 5th edn. Oxford: Blackwell Scientific Publication
Williamson JR, Tilton RG, Chang K, Kilo C (1998) Basement membrane abnormalities in diabetes mellitus: Relationship to clinical microangiopathy. Diabetes/Metabolism Reviews 4(4): 339-70
RELATED ARTICLE: ARTICLE POINTS
1 Diabetic foot wounds are a serious problem in medical, social and economic terms.
2 Thickening of the basement membrane and reduced maximal hyperaemia of capillaries in diabetics are common observations.
3 It is not known whether external pressure or vasospasm cause microvascular disease in the diabetic foot..
4 It has yet to be determined whether microvascular disease is the underlying element accountable for the development of diabetic foot ulceration.
* Diabetic foot wounds
* Microvascular disease
* Basement membrane thickening
* Reducation of maximal hyperaemia
* Microvascular pressure
1 Although its is not fully understood, a thickening of the basement membrane is a consistent finding in the capillaries of people with diabetes.
2 Thickening of the basement membrane is thought to be a result of the proliferative response of the vascular endothelium to chronic injury, brought on by the effects of hyperglycaemia on glucose metabolism, protein synthesis and non-enzymatic glycosylation of cellular and extracellular components.
3 Reduction in maximal hyperaemia has also been observed in people with diabetes.
1 A study found a link beteen impaired hyperaemic response and duration of diabetes, but not to long-term blood glucose control, in children.
2 Findings suggest that functional abnormalities of the microcirculation can occur independently of large vessel disease.
3 It has been suggested that true microangiopathy leading to diabetic complications, is due to reduced blood flow in nutritive capillaries.
1 Early studies demonstrated that microscopic degenerative changes occur when tissue is subject to pressure for short periods of time, but that the normal hyperaemic response more than compensates.
2 It has not been determined whether external pressure or vasospasm is a cause of microvascular disease in the foot.
3 It has been shown that insufficient microvascular perfusion can delay or inhibit healing.
1 The development of nephropathy and neuropathy is retarded after a few years of intense insulin treatment.
2 These results suggest a direct link between hyperglycaemia and the aetiology of microvascular disease.
3 Despite reservations, sufficient evidence exists to implicate the microvascular endothelium as responsible for a variety of pathophysiological phenomena observed in diabetes, but whether it is the underlying element accountable for the development of diabetic foot ulceration has yet to be determined.
1 There is little evidence to implicate microvascular changes. alone as sufficient to result in capillary closure.
2 The failure of the microcirculation is possibly more likely to be linked with increasing duration of diabetes and functional abnormalities of the microvasculature developing in parallel with neuropathy and occlusive arterial disease.
Elizabeth Mudge is Senior I Podiatrist with North Bristol NHS Trust.
COPYRIGHT 2002 S.B. Communications
COPYRIGHT 2003 Gale Group