Abstract
We hypothesized that foot infections secondary to a puncture wounds (PWs) have a worse prognosis concerning infection-related mortality, recurrence of the infection, and healing than those secondary to a chronic ulcer. We conducted a prospective study consisting of 200 patients with moderate-to-severe diabetic foot infections. The cohort consisted of 155 men (77.5%) and 45 women (22.5%). The mean age of the patients was 59 years (standard deviation 12.2). Puncture wounds were the cause of the infection in 107 patients (53.5%) and a chronic ulcer was the cause in 93 patients (46.5%). One hundred and eleven patients (55.5%) had moderate and 89 (44.5%) had severe infections. Osteomyelitis was more frequently found in chronic ulcers (71%) than in PWs (44.9%), P < .001. Cox's survival analysis using PWs as an explanatory variable showed no association with infection-related mortality (hazard ratio [HR] 1.06, 95% confidence interval [CI] 0.32-3.46, P = .92), time to recurrence of infection (HR 0.64, 95% CI 0.27-1.51, P = .30), and time to healing (HR 0.81, 95% CI 0.60-1.08, P = .15). More than half of our patients had PWs as the mechanism by which the infection occurred. These patients usually had a lower rate of osteomyelitis but required hospitalization and antibiotic therapy more frequently than patients with infected chronic ulcers. We found no difference in outcomes between the 2 groups.
Introduction
Diabetes-related foot infection is a challenging situation. Foot infections in people with diabetes occur when bacteria get inside the foot through a chronic ulcer. Another mechanism might be puncture wounds (PWs), which are associated with worse outcomes in patients with diabetes compared with patients without diabetes. 1 One survey-based study in Jamaica including 198 patients reported a prevalence of PWs of 25.8%. 2 Walking barefoot has been reported as a risk factor for foot disease in people with diabetes. 3 Many patients with diabetes will be unaware of the penetrating injury because peripheral neuropathy produces insensitivity to pain. This way the patient may be delaying medical assistance. On the other hand, the inflammatory markers are lower in diabetic patients with foot infections as compared with nondiabetic patients. 4 That attenuated inflammatory response could facilitate the spreading of the infection through the compartments of the foot. However, neuropathy and diabetes-related immunity disturbances are also present in chronic ulcers. Some studies have addressed the differences in the outcomes of PWs in patients with and without diabetes.4,5 However, to the best of our knowledge, no studies have compared the outcomes of foot infections secondary to PWs and chronic ulcers in patients with diabetes. We hypothesized that foot infections secondary to PWs have a worse prognosis than those secondary to a chronic ulcer.
Material and Methods
We conducted a prospective study consisting of 200 patients with moderate-to-severe diabetic foot infections (DFIs) according to the International Working Group on Diabetic Foot (IWGDF) 6 consecutively recruited from our diabetic foot department in Hospital San Juan de Dios, San Jose, Costa Rica, from October 15, 2020, to December 15, 2021. The diagnosis of infection was based on the inflammatory signs and symptoms as stated in the guidelines of the IWGDF. 6 The severity of the infection, mild, moderate, and severe was also diagnosed as suggested by the IWGDF guidelines. 6 Diabetic foot infections deeper than subcutaneous tissue with erythema >2 cm without systemic inflammatory response syndrome (SIRS) were classified as moderate. Diabetic foot infection was classified as severe when presenting 2 or more signs of SIRS. 6 Mild infections were excluded from the present study. Medical background, laboratory tests, surgical procedures, and additional information collected during follow-up were included in the research database.
This study, conducted following the Declaration of Helsinki as revised in 2008, was approved by the Ethics Committee of San Juan de Dios Hospital (Caja Costarricense del Seguro Social), San José, Costa Rica (HSJD-CEC-14-2020).
Patients with hemodynamic and/or metabolic instability, changes in mental status, extensive necrosis, or those with ischemia requiring revascularization were hospitalized. Patients who were not admitted were operated on as outpatients.
Peripheral arterial disease of the involved foot was established in cases in which one or both (dorsalis pedis and posterior tibial) foot pulses were absent, and/or the ankle/brachial index was <0.9 or >1.2. The diagnosis was always assured by a vascular surgeon.
A definitive diagnosis of osteomyelitis (OM) was only made when the surgeon visually confirmed signs of bone destruction during surgery and/or following microbiological and/or pathological confirmation of bone involvement.
Conservative surgery was defined as any surgical procedure in which the infected tissue, including the bone when required, was debrided without performing any amputation. Minor amputation includes any amputation below the ankle joint; major amputation includes those through or above the ankle joint. Surgical wounds were left open to heal by secondary intention in conservative surgery and minor amputations. Wound care was provided according to our current protocol. Offloading was provided using a removable cast walker (Optima Diab®, Molliter), crutches, wheelchairs, or bed rest depending on the characteristic of the patient and the wound.
Cases in which the surgeon confirmed that all the infected tissue (both soft tissue and bone) had been removed were treated without postoperative antibiotics. The rest of the cases were treated with empirical antibiotics until the results of microbiological samples were available. In such cases, antibiotics were modified according to the antibiogram and were stopped when the signs of infections had disappeared.
Tissue biopsies under sterile conditions were taken in the operating theatre. Anaerobic cultures were not available in our laboratory. Multidrug-resistant organisms (MDROs) include the following bacterial species: Methicillin-resistant Staphylococcus aureus, vancomycin-resistant Enterococcus, Enterobacteriaceae resistant to third-generation cephalosporins and/or carbapenem, and all antimicrobial susceptibility phenotypes of Pseudomonas aeruginosa and Acinetobacter baumannii. 7
Prognostic variables are infection-related mortality, recurrence, and healing. Infection-related mortality is defined as any death occurring in-hospital, within 30 days after discharge from the hospital, and those occurring at any time during the follow-up whereby the foot infection was established as the cause of the death.
Healing was defined as stable epithelization of the surgical wound and the ulcer that was the point-of-entry of the infection for 3 months. Recurrence of the infection was defined as any condition in which clinical signs of infection reappeared during follow-up, including impairment of the healing course and/or cases that needed additional antibiotic therapy and/or surgical revision.
Number of patients treated and cause of infection.
Statistical Methods
We used the Shapiro-Wilk test to determine normal distribution of continuous variables. We describe normally distributed continuous variables as the mean and standard deviation (SD) and non-normally distributed variables as the median and interquartile range (IQR). We present categorical variables as a number and percentage of the sample. We compared continuous variables between the groups by using Student t test (normally distributed variables) and the Mann-Whitney U test (non-normally distributed variables). We compared categorical variables between the groups by using the χ2 test and Fisher exact test when indicated. We performed univariate Cox survival analysis to study the association between infection-related mortality, time to recurrence of the infection, and time-to-healing with the variable PWs. We set P < .05 (2-tailed) as the threshold for statistical significance. We used the jamovi software for macOS for statistical analysis [The jamovi project (2021). jamovi (Version 1.6) Computer Software. Retrieved from https://www.jamovi.org] for macOS.
Results
The cohort consisted of 155 men (77.5%) and 45 women (22.5%). The mean age of the cohort was 59 years (SD 12.2). One hundred and eleven patients (55.5%) had moderate and 89 (44.5%) had severe infections. The median duration of diabetes was 14 years (IQR 12). Glycemic control was poor with a mean glycated hemoglobin (HbA1c) of 9.8% (SD 2.3%) or 83.8 mmol/mol (SD 25.5 mmol/mol). Puncture wounds were the cause of the infection in 107 patients (53.5%) and a chronic ulcer was the cause in 93 patients (46.5%) (Figure 1). Comparison between the groups is shown in Table 1. The 5 most isolated species from the 178 positive cultures were: Staphylococcus aureus, n = 52 (29.2%); Enterococcus faecalis, n = 36 (20.2%); Proteus mirabilis, n = 34 (19.1%); Escherichia coli n = 29 (16.3%); and Pseudomonas aeruginosa, n = 20 (11.2%). Multidrug-resistant organisms were isolated in 44 samples (24.6%). The microbiology of infection and inflammatory markers regarding the type of wound is shown in Table 2. Arresting of the infection was achieved by means of conservative surgery in 117 patients (58.5%), 69 minor amputations (34.5%), and 14 major amputations (7%). In cases of OM, 71 patients (62.3%) underwent conservative surgery, 34 (29.8%) minor amputations, and 9 (7.9%) major amputations. Eleven patients died of infection.
Demographics, Foot-Related History, and Management of the Infection.
Abbreviations: IQR, interquartile range; SD, standard deviation.
Inflammatory Markers and Microbiology Results From 178 Samples With Positive Bacterial Isolation.
Abbreviations: IQR, interquartile range; MDRO, multidrug-resistant organism.
Cox survival analysis using PWs as an explanatory variable showed no association with infection-related mortality (hazard ratio [HR] 1.06, 95% confidence interval [CI] 0.32-3.46, P = .92), time to recurrence of the infection (HR 0.64, 95% CI 0.27-1.51, P = .30), and time to healing (HR 0.81, 95% CI 0.60-1.08, P = .15).
Discussion
Patients with PWs were younger, had higher body mass index, worse metabolic control, and a history of fewer foot-related complications. No differences in microbiological isolates and inflammatory markers, with the exception of white blood cell count, were also found.
The median time (15 days) before our patients with PWs presented in our department is shorter than that previous series (mean 20.1 days). 1 Wound duration was shorter in cases of PWs than in chronic ulcers. It means that patients sought medical assistance sooner due to the acuteness of the infection. White blood cell count, a marker of acute infection, was also higher in the PWs group. However, no difference was found regarding the rest of the inflammatory markers evaluated in the present study.
Every patient in the present series underwent surgery, a finding that is consistent with 95% reported in another study. 1 However, that does not mean that every patient with diabetes and PWs requires debridement. In a prior study, up to 45.4% of 77 episodes healed without medical assistance because they were noninfected. 2 However, we recommend observing people with diabetes and PWs.
We have not found differences in microorganisms isolated from tissue biopsies between the 2 groups. It could be thought that MDRO would be more frequent in chronic ulcers. Indeed, diabetic foot ulcers with a longer course were reported to be a risk factor for MDRO infection. 8 No differences regarding the 5 most frequent isolates were found between the 2 groups.
Patients with PWs had a longer duration of antibiotic treatment. However, no difference in the severity of the infection between the 2 groups was demonstrated. We think that it is related to the differences in the hospitalization rate between the 2 groups. Patients with PWs required hospitalizations at a higher rate than patients with chronic ulcers. Patients requiring admission usually required a longer duration of antibiotic treatment than those treated as outpatients.
Ulcers became complicated with OM more frequently than PWs. In one case series, the authors found OM in 37.3% of patients with diabetes and PWs. 1 Our cohort has a higher rate of OM complicating PWs. However, chronic ulcers have a significantly higher rate of bone involvement compared with PWs. Osteomyelitis has been reported as a factor for the recurrence of diabetic foot ulcers. 9 One series reported that 56.7% of patients with OM had reinfections compared to 38% of those with STIs, P < .01. 10 However, PWs were not associated with the recurrence of the infection. It could be due to half of the patients with PWs just having soft tissue infections.
Our study has some strengths, it is a prospective study from a reference center dealing with DFIs in Costa Rica. The diagnosis of OM was confirmed by microbiology and pathology. Patients were properly followed up except for 5 who did not attend our consultation. Our study provides useful information for further studies dealing with PWs in people with diabetes. Criteria to decide on surgery and/or antibiotics when the patient is early evaluated or the best diagnostic approach to OM in cases of PWs are necessary. However, there are several limitations. We did not recruit patients with mild infections and some of these patients could have healed without surgery. Maybe early antibiotic treatment and offloading could prevent the worsening of the infection. Anaerobic cultures are lacking, meaning we have an incomplete view of microbiology. Maybe anaerobic bacteria are more frequently associated with these closed lesions. Every patient underwent surgery, and our outcomes may not be replicated in other clinical settings where patients are mainly treated with antibiotics. The criteria to decide whether a patient required postoperative antibiotics were based on the clinical evaluation of the surgeon; however, we did not apply other criteria, that is, inflammatory markers, and that information could have been useful.
Conclusions
More than half of our patients had PWs as the mechanism by which the infection occurred. These patients usually had a lower rate of OM but required hospitalization and antibiotic therapy more frequently than patients with infected chronic ulcers. We found no difference in outcomes between the 2 groups.
Footnotes
Author Contributions
JA-S and GV-M conceived the concept. JA-S, GV-M, and JMR-B designed the clinical database and collected data. JA-S and GV-M analyzed the data and wrote the main manuscript. JA-S, GV-M, MEL-V, JA-H, CA-H, and JMR-B reviewed the manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
