A Retrospective Assessment of Partial Calcanectomies and Factors Influencing Postoperative Course

A Retrospective Assessment of Partial Calcanectomies and Factors Influencing Postoperative Course
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   A Retrospective Assessment of PartialCalcanectomies and Factors InfluencingPostoperative Course Jeremy Cook, DPM, 1 Emily Cook, DPM, 2  Adam S. Landsman, DPM, PhD, 3 Philip Basile, DPM, 4 Thanh Dinh, DPM, 5 Thomas Lyons, DPM, 6 Barry Rosenblum, DPM, 7 and John Giurini, DPM 8 Partial calcanectomies are a common procedure for the treatment of chronic heel ulcers. We reviewed 50 cases from patients who had partial calcanectomies to determine what factors, if any, affect the rateof healing. Each case was followed up for 1 to 6 years after surgery. We found that calcanectomy woundswere difficult to heal, regardless of the etiology. We examined a multitude of factors to determine whichones affected the rate of closure, including body mass index, vascular status, preoperative albumin levels, wound grade, presence of methacillin-resistant   Staphylococcus aureus  , and other factors. Wefound that the average total closure rate for patients undergoing partial calcanectomy was between 51% and 83% after 1 year, depending on various preoperative conditions. Our data suggest that even thoughthe procedure itself is fairly straightforward, the course of recovery is complex, with a fairly high rate of failure after 1 year. We found that preoperative infections with methacillin-resistant   Staphylococcusaureus  , vascular disease, albumin levels, and preoperative ulcer grade had a significant bearing on theoutcomes. We also identified other factors that did not seem to affect surgical outcomes. Based on thesefindings, the authors suggest some factors that should be considered when performing partial calcane-ctomies.  (The Journal of Foot & Ankle Surgery 46(4):248–255, 2007)Key words: partial calcanectomy, limb salvage, heel ulcers, osteomyelitis, MRSA  I t is estimated that there are 17.7 million patients withdiabetes in the United States, and it has been projected thatthis number will double by the year 2030 (1). In 1996, the lifetime risk of a patient with diabetes developing a footulceration was 15% and has now been estimated to be ashigh as 25% (2). Heel ulcers have been reported to be themost serious type of foot ulceration leading to a proximalamputation (3). The National Pressure Ulcer Prevalence Survey of 1995 found that of the 39,874 patients in 265hospitals, the incidence of heel ulcers ranged from 19% to32% (4). Nonhealing ulcers have been linked to 85% of  major amputations (5). Long-term survival rates of major amputations in patients with concurrent diabetes or renaldisease have been found to be dismal, with a 70% 1-yearsurvival rate and only 30% 5-year survival rate in a recentoutcomes series (6). Therefore, limb salvage is of para-mount importance for surgeons who treat this affliction.Originally presented in 1931 (7) but described as early as 1896 (8), the partial calcanectomy (PC) has been an estab- lished limb salvage procedure that offers patients an alter-native to amputation. Clinically, PCs are frequently chosenfor patients with substantial soft tissue loss, and may in-volve extensive bony exposure. Figure 1 demonstrates asevere ulceration of the heel, which required a PC. Figure 2demonstrates the marked loss of soft tissues, around thecalcaneus, apparent on this radiograph. Figures 3 and 4 show the clinical and radiographic postoperative appearanceof the foot after PC.The largest study published to date used PC in 29 patients Address correspondence to: Adam Landsman, DPM, PhD, FACFAS,Beth Israel Deaconess Medical Center, Department of Surgery, HarvardMedical School, One Deaconess Road, Boston, MA 02215. E-mail:alandsma@bidmc.harvard.edu. 1 Chief Resident, Beth Israel Deaconess Medical Center, Department of Surgery, Harvard Medical School, Boston, MA. 2 Chief Resident, Beth Israel Deaconess Medical Center, Department of Surgery, Harvard Medical School, Boston, MA. 3 Beth Israel Deaconess Medical Center, Assistant Professor, Depart-ment of Surgery, Harvard Medical School, Boston, MA. 4 Beth Israel Deaconess Medical Center, Clinical Instructor in Surgery,Department of Surgery, Harvard Medical School, Boston, MA. 5 Beth Israel Deaconess Medical Center, Clinical Instructor in Sur-gery, Department of Surgery, Harvard Medical School, Boston, MA. 6 Beth Israel Deaconess Medical Center, Clinical Instructor in Surgery,Department of Surgery, Harvard Medical School, Boston, MA. 7 Beth Israel Deaconess Medical Center, Assistant Professor andDirector of Resident Training, Department of Surgery, Harvard MedicalSchool, Boston, MA. 8 Beth Israel Deaconess Medical Center, Associate Professor and Divi-sion Chief of Podiatric Surgery, Department of Surgery, Harvard MedicalSchool, Boston, MA.Copyright © 2007 by the American College of Foot and Ankle Surgeons1067-2516/07/4604-0008$32.00/0doi:10.1053/j.jfas.2007.03.016 248  THE JOURNAL OF FOOT & ANKLE SURGERY  over a 10-year period for the treatment of heel ulceration. Inthis study, of the diabetic subgroup, which consisted of 18patients, only 22% of these patients went on to heal ascompared with 78% of the nondiabetic group. The finalhealing endpoint demonstrated a 65% failure rate in patientswith diabetes (9). Other studies have reported healing rates of 40% to 100% (10–15), with the majority having a smallsample size or consisting only of case studies (16–21).Many studies do not define the amount of calcaneus re-sected (10, 13, 22) or presence of osteomyelitis (9, 11, 14), include pediatric patients (23, 24), or focus on traumatic (11, 25) or hematogenous etiology (23, 26). The purpose of the current study was to evaluate theoutcomes of PCs in a large, consecutive series of nontrau-matic but limb-threatening heel ulcerations. This study alsosought to delineate associated preoperative conditions thatimpacted these outcomes. The complex nature of thesewounds, including their associated comorbidities, promptedour team to hypothesize that several factors may delay therate of healing after PCs. Among these elements with as-sumed predictive value were the presence of peripheralvascular disease (PVD), body mass index (BMI), poor nu-tritional status determined by albumin levels, wound classas described by the Wagner system (27), and the presence of methacillin-resistant  Staphylococcus aureus  (MRSA). Materials and Methods A total of 50 feet from 46 patients who underwent a PCbetween 2000 and 2004 at the Beth Israel Deaconess MedicalCenter,Boston,Massachusetts,wereincludedinthisretrospec-tive study. Fifteen of these patients were patients with type 1diabetes, 31 were patients with type 2 diabetes, and 1 was apatient without diabetes. In all cases, at least 1 year of fol-low-up was available, and, in some cases, as much as 6 yearsof follow-up was considered with a mean of 2.7 years of follow-up available. This study focused on the postoperativecourse of healing based on the preoperative health status of thepatient. As part of our assessment, we reviewed the charts togather the following information for analysis: ●  Duration of ulcer before surgery ●  Age of patient ●  History of tobacco use ●  Preoperative ulcer grade ●  Albumin levels ●  Presence of MRSA versus other bacteria ●  BMI ●  Vascular disease and need for interventionUsing these data points, we evaluated what affect, if any,they had on the ability of a patient to heal his or her wounds, FIGURE 1  This patient presented with extensive soft tissue lossand exposed calcaneus. PC, at the very least, would be necessaryto close this type of wound. FIGURE 2  This radiograph illustrates the extent of soft tissue lossas well as the condition of the calcaneus bone before surgicalintervention.  VOLUME 46, NUMBER 4, JULY/AUGUST 2007  249  and the amount of time necessary to achieve closure. Wealso included the rate of postoperative dehiscence afterprimary closure in our review.For the purposes of this study, a wound was consideredhealed if there was full epithelialization over the entire heelwithout further breakdown, and absence of drainage. Al-though we tracked wound sites for up to 6 years, the focusof our analysis was during the first year after PC surgery.Analysis was performed with the Student  t   test and analysisof variance for parametric data, and chi-square analysis wasused for nonparametric data. In addition, averages and stan-dard deviations (SD) were used to characterize progressionof healing in some cases.Preoperative grading of the ulcer was based on the Wag-ner classification system (27) and was recorded for each participant. Serum albumin was collected during the asso-ciated admission before the calcanectomy and was considernutritionally deficient if levels were noted to be  3.0 g/dL.BMIs were prepared by using measurements of weight andheight performed in the hospital. Bacterial isolates wereidentified from intraoperative specimens submitted for mi-crobial culture and gram staining. Patient data points such asage, history of tobacco, and duration of ulceration were self reported. Results The data presented here were gathered retrospectivelyfrom our medical records. We analyzed data from 50 PCsperformed on 46 patients. The average age was 56.7 years(SD  10.2 years), with 32 men and 14 women evaluated.Seven patients were lost to follow-up immediately aftersurgery, and 1 patient died 6 days after surgery from acerebrovascular accident.In our study, we found that complete healing wasachieved in 71.4% during the first year, and this numberincreased to 76.2% when looking beyond the first year FIGURE 3  This picture demonstrates a completely healed patient after PC. The posterior defect is apparent, as is the total closure of thewound site. FIGURE4  ThisradiographisofthesamepatientasinFigure3anddemonstrates the extent of bony resection that was necessary toachieve closure. 250  THE JOURNAL OF FOOT & ANKLE SURGERY  follow-up period. The average time to full closure was201.2 days (SD    282.7 days). Although many of thewounds were clinically closed at the time of surgery, ourcriteria for healing required removal of all sutures and nogaps or dehiscence noted along the incision lines. There wastremendous variation in the number of days required fortotal closure based on these criteria, with a range of 27 to1532 days observed.Based on our criteria for vascular disease (nonpalpablepulses and ankle/brachial index  0.7), 28 patients requiredsome form of vascular intervention before surgery. Inter-vention involved angioplasty, with or without insertion of astent, and/or distal bypass surgery. We found that the pa-tients with vascular disease who underwent an interven-tional procedure closed 65.2% (n  23), whereas those whodid not have vascular disease closed 68.4.0% (n    19) of the time, within 1 year. The average time to closure waslonger with the patients with vascular disease who required120.5 days (SD    74.78 days) as compared with the pa-tients without vascular disease, who required 106.3 days toclose (SD    89.7 days). Although a difference in healingtime was noted, it was not statistically significant ( P  .33).This trend becomes slightly stronger when consideringlonger-term closure rates. We found that 73.9% of thewounds in patients with treated vascular disease closedeventually, whereas 84.2% closed in patients without vas-cular disease ( P  .22).Using the scale supported by the American Society forBariatric Surgery (28, 29), we found that the rate of  healing was not measurably affected by the patient’s BMI(Table 1). As with all open wounds, the risk of bacterial infection isquite high. In our study, 48 out of 50 (96%) of the woundsgrew out bacteria. Of these, 19 (38.8%) had MRSA isolates.The 35 patients who produced nonMRSA colonies werenoted to have a 71.4% closure rate as compared with 50.0%in the MRSA subgroup during the first-year period ( P   .09). When we compared the time to healing between these2 categories, we found that the nonMRSA patients healed,on average, in 86.6 days (SD  50.9 days), whereas thosewith MRSA isolates averaged 335.7 days (SD    370.0days); the difference between these 2 averages proved to bequite significant ( P  .004).To assess the affect of nutritional status, preoperativeserum albumin levels were evaluated. Data were availablefor 32 of the patients in our study. We found that 56.3% of the patients with a low preoperative albumin level of 2.2 to3.0 g/dL (n    16) healed during the first year, whereaspatients with an albumin level of   3.0 g/dL healed 68.8%of the time, during that same postoperative interval. Theaverage time to heal in the deficient group was 160.9 days(SD    84.1 days), whereas the group with albumin levels  3.0 g/dL healed in 98.5 days (SD  77.4 days) during thefirst year after surgery. This difference in healing time wasstatistically significant ( P  .05).Preoperative wound classification was one parameter thatstrongly influenced the rate of wound closure. Our studyincluded ulcers that ranged from Wagner grades 1 to 4, withthe vast majority of the wounds being grade 2 or grade 3.Analysis demonstrated that 83.3% of the grade 2 Wagnerwounds closed within 1 year, whereas only 51.5% of thegrade 3 Wagner wounds closed in the same time period.This difference in closure rate was statistically significant( P    .03). Wagner grade 2 ulcers typically required 97.5days (SD  45.0 days) to close, as compared with grade 3ulcers, which required 142.6 days (SD  85.8 days).We also considered whether the length of time that theulcer was present before surgery (that is, wound age) hadany influence on the rate of healing and found that there wasno correlation between the age of the ulcer and the potentialfor healing. In fact, some of the oldest ulcers responded thequickest to surgical intervention (Fig 5). Patients with a past medical history of smoking showedabout the same closure rate, but took a longer time to closetheir wounds as compared with nonsmokers. Nonsmokersrequired approximately 177.3 days (SD    130.4 days),whereas smokers required 235.3 days (SD  381.6 days) toachieve closure; however, the difference was not statisti-cally significant because of the large SD observed, particu-larly among smokers.Another point of consideration was the impact of thepatient’s vascular status on the rate of dehiscence. Weconsidered whether our patients who were vascularly com-promised and underwent preoperative intervention (that is,angioplasty, stent, or bypass) were more likely to dehisceafter primary wound closure. Our results demonstrated that88.2% of the patients who were vascularly compromiseddehisced despite vascular intervention, whereas only 35.7%of those patients with adequate circulation developed thiscomplication. This difference was statistically significant( P  .001). We also found that 40% of the patients in ourstudy experienced a recurrent ulcer at some point, after totalwound closure.Of the 46 patients monitored in our study, 8 ultimatelywent on to a below-the-knee amputation (BKA) secondaryto their initial ulceration, 2 underwent BKA for an unrelatedcondition, and 2 had contralateral BKAs. One of the 8 BKA TABLE 1 BMI versus wound healing BMI Wounds Healed by1 Year Average Time toHeal (d)  18.4 (underweight) 66.7% 139.018.5–24.9 (normal) 43.0% 112.025.0–29.9 (overweight) 66.7% 119.330–34.9 (obese, class 1) 66.7% 62.335.0–39.9 (obese, class 2) 66.7% 204.0  40.0 (morbidly obese) 75% 135.0  VOLUME 46, NUMBER 4, JULY/AUGUST 2007  251  cases was performed bilaterally and is in addition to thecontralateral BKAs reported. Six patients were lost to fol-low-up. A Kaplan-Meier Life Table analysis (30) was per- formed (Table 2), and this reveals a probability of survivalof 0.883 during the first 4 years after surgery. Discussion In this study, we considered many different parametersthat could affect the outcome of PCs. Although the focus of our analysis was closure of the wound, there are otherconsiderations as well, which are not measured here. Forexample, we did not consider whether our patients couldambulate after surgery, or what type of braces or shoe gearwould be required. Similarly, we plan to discuss the influ-ence of adjunctive treatments, such as skin grafting and thewound vacuum assisted closure, in future publications.In our study, closure of a wound was the critical endpointin our analysis. All parameters were discussed in the contextof whether this endpoint was achieved. In fact, we foundthat there were several factors that influenced a patient’sability to achieve closure, as well as the time necessary forsuch closure.Several factors that had a clear bearing on this outcomewere identified. We found that the presence of MRSAappears to have a strong influence on the rate of woundclosure. Patients infected with MRSA required more time toachieve closure. In a study by Dang et al (31), at least 30.2% of the diabetic foot infections in their series of 93 patientswere MRSA isolates. Nearly all of our patients demon-strated preoperative bacterial infection, and 38.8% hadMRSA isolates. Although the prevalence of   Staphylococcusaureus  as a pathologic entity is well established (31), our review of the literature failed to produce any evidencethat MRSA is associated with prolonged surgical healing.Our results demonstrated a greater 1-year closure rate inthe nonMRSA arm (71.4%) versus only 50% in thosepatients with MRSA isolates ( P    .004). Given ouraverage difference of 249.1 days in healing time betweenthose with MRSA isolates and those with nonMRSA FIGURE 5  This graph shows the correlation between the age of the ulcer and the time required for healing, after PC. TABLE 2 Kaplan-Meier Life Table analysis (number in trial  49) Year Number Died Number Censored Survival Probability 95% Confidence Interval Number Surviving1 3 6 0.93 1.009 to 0.851 402 1 0.906 0.998 to 0.815 393 1 0.883 0.985 to 0.781 384 0 0 0.883 0.985 to 0.781 38 252  THE JOURNAL OF FOOT & ANKLE SURGERY
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks