predictors.30 Foot deformity did not enter the final analytic model in the studies by Litzelman et al.49 or Kastenbauer et al.37 A number of other studies not presented in Table 1.4 have assessed the role of elevated foot pressure in foot ulcer development, using case, control com- parisons.
As long ago as 1963, Bauman and Brand found elevated plantar pressures under the feet of people with neuropathic insensitivity, foot deformities and foot ulcers.52 Mueller and colleagues found that patients with diabetes and a history of foot ulcers had significantly reduced ankle dorsiflexion and subtalar joint range of motion, compared to those without diabetes.53 In a similar study, Zimny et al. found that those with diabetic neuropathy but no history of foot ulcers also had reduced dorsiflexion and subtalar motion, compared to non- diabetic controls.54 Recently, Robertson et al. found, using spiral computerised tomography, that plantar tissue muscle density was decreased and that metatarso-phalangeal arthropathy (especially hammer toe deformity) was more likely to be present in those with diabetic periph- eral neuropathy and a history of plantar ulcer than in normal controls (P < 0.001).55 In another study by the same group, Mueller et al. found that peak plantar pressure during walking was significantly greater in those who had both diabetic peripheral neuropathy and hammer toes than in normal controls.56 Van Schie et al. found a greater frequency of both foot deformities (hammer toes, claw toes, prominent metatarsal heads and high medial arch) and foot muscle weakness (in both intrinsic and extrinsic muscles) (P < 0.001, Kruskal, Wallis test for trend in both types of comparisons) in those with a history of diabetic foot ulcers than in diabetic, non-neuropathic and non-diabetic controls.57 In reports from prospective cohort studies, progressively higher plantar pressure predicts increasing foot ulcer risk.35,58 Despite these associations, prospective studies have been unable to demonstrate an optimal cut-point for increasing plantar pressure, above which the probability of foot ulceration is substantially increased.43,59 This may be the case because other factors, such as weight-bearing activity, act together with plantar pressure to increase foot ulcer risk.
Maluf and Mueller in a case, control study found that cumulative plantar tissue stress (which they defined as the combination of plantar pressure and total daily weight-bearing activity) was reduced in those with a history of diabetic neuropathic ulcers compared to either those with neuropathy alone or non-diabetic controls (P = 0.03).60 The authors speculated that plantar tissues in those who ulcerated may have been more vulnerable to ulceration due to disuse atrophy.
That study measured plantar pressure once at study onset, and measured weight- bearing activity over the ensuing week.
Ledoux et al. investigated the relationship between ulcer location and peak plantar pressure at one Veterans Affairs (VA) medical centre in 549 individuals with diabetes, each of whom had in-shoe plantar pressure measured using the F-scan plantar pressure measurement device.
After an average of 2.5 years of follow-up, there were 42 patients who developed plantar ulcers.
In an analysis that considered whether plantar pressure differed within each foot site by foot ulcer occurrence, no significant difference was seen for peak pressure.
Sites at which ulcers developed had higher mean pressure than other sites, but the site of highest pressure was unrelated to the foot ulcer site.61 Together, these studies represent a substantial shift over time in our understanding of the role of plantar pressure and foot ulceration.
It is becoming clear that while foot deformities and associated plantar pressure are important risk factors for diabetic foot ulcer, other as yet unidentified factors probably play an important synergistic role with plantar pressure in the development of foot ulcers.
Further prospective studies are needed to investigate the joint role of plantar pressure and weight-bearing activity using technology that measures cumulative plantar tissue stress continuously, via an in-shoe system.
Such systems are being developed62 and will greatly improve investigation in this area.
JWBK089-01 JWBK089-Boulton April 21, 2006 20:30 Char Count= 0
ULCER OUTCOMES 11
ULCER OUTCOMES Table 1.5 shows the outcomes of incident ulcers reported from three prospective studies of foot ulcers.
The proportion of patients with diabetic foot ulcer who progress to at least par- tial amputation ranges from 11 to 24%, depending on ulcer severity and length of follow- up.26,28,46,47,63 Factors associated with amputation once foot ulcer occurs will be reviewed in the next chapter.
Later chapters will review therapeutic strategies for treatment of dia- betic foot ulcer; however, a number of studies have found that, given similar care, ulcer surface area and ulcer duration prior to the start of treatment delay ulcer healing.13,63, 66 In a study of 194 ulcers that were re-examined weekly for 6, 18 months, Oyibo and colleagues found that ulcer surface area differed strongly and significantly between ulcers that healed, did not heal or proceeded to amputation (larger ulcers having worse outcomes and taking longer to heal).
Patient gender, age and duration of diabetes at presentation, and site of the ulcer on the foot did not affect time to healing.
Neuroischaemic ulcers took longer to heal (20 vs 9 weeks) and were three times more likely to lead to amputation.63 Margolis and colleagues found, after pooling data from the control arms of five related randomised stud- ies investigating new ulcer-healing therapies, that neuropathic wounds were more likely to heal within 20 weeks if they were smaller (<2 cm2 ), had existed for a shorter period be- fore they were treated (<6 months) or if the patients were of non-White ethnicity.13 Gen- der, age and glycosylated haemoglobin level had no effect in their multivariable regres- sion model.
In an analysis that utilised medical records from 150 wound care facilities in 38 US states, these same investigators confirmed that among 72 525 diabetic foot wounds in 31 106 patients, wounds that were older, larger and deeper in grade (especially Wagner grade ≥ 3) were more likely to take more than 20 weeks to heal, after adjustment for gender and age.64 Pecoraro and colleagues described the importance of a 4-week reduction in ul- cer volume and reported that low levels of periwound TcpO2 and CO2 were significantly associated with initial rate of healing, while an average periwound TcpO2 <20 mm Hg was associated with a 39-fold increased risk of early healing failure.66 Sheehan and col- leagues similarly found, among 276 patients with Wagner grade ≥ 1 diabetic foot ulcers of 30 days duration, that change in ulcer area within 4 weeks of treatment onset strongly predicted complete wound healing by 12 weeks.65 All patients in each of these studies re- ceived similar ulcer care, which included offloading, wound debridement and moist wound dressings.13,63, 65 Foot ulcer recurrences were addressed in a UK study by Mantey and colleagues.67 Diabetic patients with an initial foot ulcer and two ulcer recurrences were compared with diabetic patients who had only one ulcer and no recurrences over a 2-year interval.
The authors reported greater peripheral sensory neuropathy and poor diabetes control in the ulcer recurrence group.
Members of the ulcer recurrence group had higher glycosylated haemoglobin levels, waited longer after observing a serious foot problem until seeking care and consumed more alcohol than did the group without ulcer recurrences.
Several years later, Connor and Mahdi reported on their cohort analysis of 83 patients followed up for 2, 10 years after their initial foot ulcer.68 They found that the 37% of patients with a higher rate of recurrence (≥3.5 ulcers per foot per 10 years) accounted for 68% of all inpatient days and 75% of all amputations.
These patients fell into two distinct groups: those with neuroarthropathy, who were more likely to wear non-orthotic footwear and had problems with footwear or orthoses, and those without neuroarthropathy, who attended clinic irregularly.
Both groups had poorer glycaemic control than those with less ulcer recurrence.
JWBK089-01 JWBK089-Boulton April 21, 2006 20:30 Char Count= 0
12 EPIDEMIOLOGY AND ECONOMIC IMPACT OF FOOT ULCERS
Table 1.5 Frequency of lesion outcomes for diabetic foot ulcers in three prospective studies
Most severe lesion All lesions followed All lesions followed followed until final until final outcomea for 6, 18 monthsb outcomec
Number of ulcers 314 194 302 Re-epithelialisation/primary 63 65 81 healing (%) Amputation at any level (%) 24 15 14 Remained unhealed (%) 0 16 0 Death (%) 13 3.5 5 Total (%) 100 100 100
a Ref. 46 (Apelqvist et al.); lesions were characterised according to Wagner criteria from superficial non-necrotic to major gangrene. b Ref. 63 (Oyibo et al.); lesions were grade 1 or deeper in the S(AD) SAD foot ulcer classification system. c Ref. 47 (Reiber et al.); study patients were enrolled with a lesion through the dermis that could extend to deeper tissue.