Abstract
Background:
Surgical site infection is one of the most common complications following ankle fracture surgery. These infections are associated with substantial morbidity and lead to increased resource utilization. Identification of risk factors is crucial for developing strategies to prevent these complications.
Methods:
We performed an age and sex-matched case-control study to identify patient and surgery-related risk factors for deep surgical site infection following operative ankle fracture treatment. We identified 1923 ankle fracture operations performed in 1915 patients from 2006 through 2009. A total of 131 patients with deep infection were identified and compared with an equal number of uninfected control patients. Risk factors for infection were determined with use of conditional logistic regression analysis.
Results:
The incidence of deep infection was 6.8%. Univariate analysis showed diabetes (odds ratio [OR] = 2.2, 95% confidence interval [CI] = 1.0, 4.9), alcohol abuse (OR = 3.8, 95% CI = 1.6, 9.4), fracture-dislocation (OR = 2.0, 95% CI = 1.2, 3.5), and soft-tissue injury (a Tscherne grade of ≥1) (OR = 2.6, 95% CI = 1.3, 5.3) to be significant patient-related risk factors for infection. Surgery-related risk factors were suboptimal timing of prophylactic antibiotics (OR = 1.9, 95% CI = 1.0, 3.4), difficulties encountered during surgery, (OR = 2.1, 95% CI = 1.1, 4.0), wound complications (OR = 4.8, 95% CI = 1.6, 14.0), and fracture malreduction (OR = 3.4, 95% CI = 1.3, 9.2). Independent risk factors for infection identified by multivariable analyses were tobacco use (OR = 3.7, 95% CI = 1.6, 8.5) and a duration of surgery of more than ninety minutes (OR = 2.5, 95% CI = 1.1, 5.7). Cast application in the operating room was independently associated with a decreased infection rate (OR = 0.4, 95% CI = 0.2, 0.8).
Conclusions:
We identified several modifiable risk factors for deep surgical site infection following operative treatment of ankle fractures.
Level of Evidence:
Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.
Surgical site infection is one of the most common complications following ankle fracture surgery. The development of surgical site infection may lead to serious consequences such as permanent disability, amputation, or death1. These infections also extend the total hospital stay and may increase health-care costs by >300%2,3. Therefore, identification of risk factors for surgical site infection is important.
The reported rate of surgical site infection following operative treatment of ankle fractures varies considerably, ranging from 1.4% to 5.5%1,4,5. Infection rates as high as 19% have been reported in diabetic patients6. This wide range of reported rates may be partly due to the inconsistent definition of superficial and deep infection. Commonly recognized patient-related risk factors for surgical site infection are diabetes4,6,7, open fracture8,9, tobacco use10, and alcohol abuse11,12. Known surgery-related risk factors are improper timing of antibiotic prophylaxis, suboptimal control of perioperative blood glucose levels, and use of a drain or nonocclusive wound dressings13-16. However, there are only limited data on surgical risk factors specifically associated with operative treatment of ankle fractures.
The aim of this case-control study was to identify patient and surgery-related risk factors predisposing patients to deep surgical site infections following operative treatment of ankle fractures. Our hypothesis was that there are previously unidentified risk factors that may affect the incidence of deep surgical site infections. Identifying modifiable risk factors, and implementing appropriate interventions, could provide the means to reduce the incidence of deep surgical site infections.
Study Design
We performed an age and sex-matched case-control study at a level-I trauma center. Approval from our institutional review board was obtained prior to the beginning of the study. All patients who had undergone surgery to treat an ankle fracture at our institution from January 2006 through December 2009 were identified by querying the hospital surgical procedure database for diagnoses coded with International Classification of Diseases, Tenth Revision (ICD-10), for fibular fracture (S82.4), medial malleolar fracture (S82.5), lateral malleolar fracture (S82.6), and bimalleolar or trimalleolar fracture (S82.8) and for procedure codes for internal or external fixation of ankle fractures. Eligible surgical procedures were restricted to those performed primarily at our institution in patients eighteen years of age or older. We identified 1923 ankle fracture operations in 1915 patients who were all definitively treated with open reduction and internal fixation (ORIF). The number of treating surgeons was ninety-three. None of the patients had a simultaneous bilateral fracture. In the control group, there were two patients who had sustained a subsequent contralateral ankle fracture during the study period but only the first fracture was studied.
A standardized operative and postoperative protocol was used during the study period. ORIF was performed according to AO principles, and the decision whether to apply a tourniquet depended on the personal preference of the treating surgeon. The wound was closed in three layers (peroneal fascia, subcutaneous layer, and skin). Postoperatively, a cast was applied in all cases either immediately in the operating room or during the following postoperative days. Sutures or staples were removed at two weeks, after which the patients were allowed to begin active ankle motion exercises. Full weight-bearing was allowed at four weeks, and the cast was removed at six weeks.
Identification of Deep Surgical Site Infection
The medical and microbiological records of all 1915 patients with an ankle fracture were reviewed for recorded signs and symptoms of surgical site infection. Two hundred and sixty-seven (13.9%) of the 1915 patients had signs of surgical site infection. We classified infections as deep when all three of the following criteria were met simultaneously: clinical signs of a surgical site infection (redness, swelling, drainage, or dehiscence), positive bacterial cultures of specimens from the wound, and osteosynthesis material visible or palpable in the wound. One hundred and thirty-one (6.8%) of the 1915 patients fulfilled the aforementioned criteria to be classified as having a deep infection. For these 131 patients with a deep infection, an age and sex-matched control group was randomly selected from a cohort of patients who had been treated operatively for an ankle fracture at our institution from January 2006 to December 2009 but did not subsequently develop a surgical site infection. The characteristics of the patients included in the study are shown in Table I.
Data Collection
Potential patient and surgery-related risk factors for deep surgical site infection were identified by a review of medical, operative, microbiological, and radiological records. The charts of each patient and records from other medical specialties (e.g., internal medicine and neurology) were also assessed. We collected the demographic data and possible comorbidities of the patients, injury mechanism (low-energy injury [defined as a same-level fall] or high-energy injury [defined as a motor-vehicle accident or a fall from a height of >1 m), delay from fracture to admission, condition of the soft tissues (Tscherne grade [0 to 4] for closed fractures and Gustilo grade [I, II, or III] for open fractures), presence of a fracture-dislocation, and fracture type (Weber classification [A, B, or C] and whether the fracture was unimalleolar, bimalleolar, or trimalleolar). The delay from admission to surgery (less than twelve hours, twelve to forty-eight hours, or more than forty-eight hours), use of a temporary external fixator, duration of surgery, use of a tourniquet, surgeon experience (specialist, or resident operating without supervision of an attending surgeon), and possible difficulties encountered during the surgery (osteoporotic bone, comminuted fracture, bone loss) were recorded. Additionally, suboptimal timing of antibiotic prophylaxis (administered more than sixty minutes before the incision or administered after the incision, or administered less than five minutes before inflation or administered after inflation of the tourniquet), wound closure method (staples or interrupted monofilament sutures), and application of a cast in the operating room were assessed. Postoperative noncompliance (defined as not adhering to the postoperative weight-bearing regimen) and wound complications (skin necrosis or blistering) were recorded, as were fracture malreduction (>2 mm in any projection) on postoperative radiographs. C-reactive protein values, blood leukocyte counts, and the causative pathogens were recorded at infection onset.
Data Analysis
An independent biostatistician performed the statistical analysis of the data. The differences in demographic and preoperative characteristics between the case and control groups were tested with use of the McNemar test (dichotomous variables) and the test of marginal homogeneity (polytomous variables). The McNemar test was also used to analyze differences between the groups with regard to postoperative noncompliance. The Mann-Whitney U test was used to compare the duration of surgery between residents and specialists, and the Kruskal-Wallis one-way analysis of variance on ranks was used to compare the operative time among unimalleolar, bimalleolar, and trimalleolar fractures. Conditional logistic regression analysis was used to determine significant risk factors for deep surgical site infection. Multivariable conditional logistic regression analysis with use of a forward stepwise procedure was applied to identify independent risk factors for deep surgical site infection. Sixteen risk factors (with p values of <0.20 in the univariate analysis) were selected at step zero, and three risk factors were included into the final multivariable model. In the final model, multicollinearity between the risk factors was not detected (tolerance of >0.97 for all three variables). Results of logistic regression analyses are expressed with use of odds ratios (OR) with their 95% confidence interval (CI). P values of <0.05 were considered significant.
Sources of Funding
There were no external funding sources for this study.
The incidence of deep surgical site infection following ankle fracture surgery was 6.8% (131 of 1923). The mean age of the patients was fifty-six years (range, twenty to ninety years), and 56% were women. The mean time from the injury to the operative fracture reduction (ORIF or temporary external fixation) was two days in both groups. The mean operative time was eighty-eight minutes (range, seventeen to 382 minutes) for the patients with deep infection and sixty-nine minutes (range, fourteen to 240 minutes) for the controls (p < 0.001). A tourniquet was not used for eighteen patients (14%) with a deep infection and for twenty-two control patients (17%). Five patients (4%) with postoperative infection had multiple concomitant risk factors (diabetic smoker with compromised soft tissue), whereas none of the controls had this risk factor combination.
In the group with deep infection, the mean duration of surgery was sixty-three minutes (range, nineteen to 149 minutes) for the patients with a unimalleolar fracture and 102 minutes (range, thirty-two to 260 minutes) for those with a trimalleolar fracture (p < 0.001). The mean duration of surgery was eighty minutes (range, seventeen to 204 minutes) for the residents and 103 minutes (range, twenty-four to 382 minutes) for the specialists (p = 0.139).
The deep infection was diagnosed at an average of 127 days after the internal fixation of the fracture. At the time of infection onset, 60% of the patients presented with an elevated C-reactive protein value (>10 mg/L). Similarly, 52% of the patients had an elevated blood leukocyte count (>8.2 E9/L). Eighty-eight (67%) of the 131 infections were monobacterial, and the three most prevalent causative pathogens in these eighty-eight infections were Staphylococcus aureus (n = 43), Staphylococcus epidermidis (n = 34) and Pseudomonas aeruginosa (n = 3). The remaining forty-three infections (33%) were multibacterial, and the most frequent pathogens were Staphylococcus epidermidis (n = 23), Staphylococcus aureus (n = 16), and Enterococcus faecalis (n = 8).
Local wound irrigation and debridement was performed, and empiric antibiotic treatment was initiated in all patients. If necessary, the antibiotic treatment was later modified according to the results of the antimicrobial sensitivity tests. Patients with a deep infection underwent an average of two additional surgical procedures (range, zero to ten). Thirty-seven patients (28%) required flap coverage, and hardware removal was performed in seventy-one patients (54%). Altogether, 103 (79%) of the 131 patients had at least one subsequent operation due to the infection. Only twenty-five patients responded to a prolonged course of antibiotic therapy and local wound care without a need for additional surgery.
Patient-related factors that were associated with a significantly increased risk of deep surgical site infection in the univariate analysis are shown in Table II. In addition to these factors, postoperative noncompliance significantly increased the risk of deep surgical site infection (thirteen patients with deep infection and no patients in the control group were noncompliant, p < 0.001). However, obesity (body mass index of >30 kg/m2) (p = 0.406), an American Society of Anesthesiologists score of 3 or 4 (p = 0.051), neuropathy (p = 0.080), schizophrenia (p = 0.054), a delay from fracture to admission of more than two days (p = 0.301), the injury mechanism (p = 0.416), the fracture type (p = 0.631), the presence of an open fracture (p = 0.280), and a delay from admission to surgery (p = 0.157) did not increase the risk of deep surgical site infection. The univariate associations of surgery-related factors and deep surgical site infection are shown in Table III. No significant associations were found with regard to tourniquet use (p = 0.494), wound closure method (p = 0.564), surgeon experience (p = 0.273), or use of a syndesmotic screw (p = 0.436).
In the multivariable analysis, the variables that remained independently associated with an increased risk of deep surgical site infection included tobacco use (OR = 3.7; 95% CI = 1.6, 8.5) and a duration of surgery of more than ninety minutes (OR = 2.5; 95% CI = 1.1, 5.7). Application of a plaster cast in the operating room (OR = 0.4; 95% CI = 0.2, 0.8) was independently associated with a decreased risk of infection.
The focus of the present study was to identify significant patient and surgery-related risk factors for deep infection following operative treatment of ankle fractures. The study identified several modifiable risk factors, including prolonged operative time, nonanatomic fracture reduction, and delayed cast application. To our knowledge, these factors have not previously been shown to increase the rate of deep infections in this setting. The results of the current study also suggest that special attention should be paid to certain patient groups, since smokers, alcohol abusers, and patients with diabetes are clearly at high risk for deep infection. Although the detrimental effects of chronic diseases cannot be resolved in the acute setting, several risk factors are modifiable and should be addressed in the operative treatment of ankle fractures. Our findings could serve as a basis for optimizing treatment algorithms for patients undergoing ankle fracture surgery, thereby reducing morbidity and associated health-care costs.
In our study, the rate of deep surgical site infection was 6.8%, which is slightly higher than the rates in previous reports1,4,5. The reported incidence of surgical site infection following ankle fracture surgery varies considerably, partially because the depth of the infection has not been clearly defined. The relatively high infection rate reported in our study may be due in part to overrepresentation of complex fractures and patients with multiple comorbidities referred from other community or smaller tertiary-care hospitals.
Although patient-related risk factors can rarely be eliminated, their effect should be minimized. In this study, diabetes was associated with an increased risk of deep surgical site infection. Uncontrolled diabetes is one of the most commonly recognized harbingers of postoperative infection4. As the incidence of diabetes increases17, so does the role of perioperative blood glucose level optimization to reduce the rate of postoperative infections18. Smoking is another modifiable risk factor; it has been found to increase the risk of surgical site infection up to fivefold following ankle fracture surgery10. Our results support this finding, as tobacco use was the strongest predictor of deep infection even after adjustment for all other variables. Therefore, every smoker undergoing ankle fracture surgery should be encouraged to quit. Even a reduction in smoking should be encouraged, since this may have beneficial effects, especially in cases with compromised soft tissue19.
A delay from admission to surgery did not increase the risk of deep infection in this current study. Previous studies have shown that, in the subgroup of patients with ankle fracture-dislocation, a delay in surgery increases the risk of postoperative infection20,21. Similarly, we noted that a fracture-dislocation or even a superficial skin abrasion increases infection risk. On the basis of these findings, judicious timing of surgery allowing for soft-tissue recovery may be warranted in cases without an associated ankle fracture-dislocation.
The goal of fracture surgery should be achieving and maintaining an anatomic reduction while minimizing the duration of surgical wound exposure to surrounding pathogens. In this study, we chose ninety minutes to be the threshold for prolonged surgery since this was reasonably close to the mean operative time in our cohort. On the basis of the multivariable analysis, prolonged operative time was an independent risk factor for deep infection. This was further corroborated by the finding that encountering intraoperative problems also significantly increased infection risk in the univariate model. Tourniquet use has been linked to postoperative pain22, but it provides a bloodless operative field, facilitating reduction and helping to shorten operative time. Although tourniquet use does not increase the infection rate following plating of tibial fractures23, it has been related to a higher incidence of postoperative wound complications after foot and ankle surgery22. In the present study, tourniquet use was not associated with an increased risk of infection. The optimal wound closure method is not known, but there is some evidence to indicate that subcutaneous wound closure with monofilament sutures minimizes tissue ischemia and is associated with decreased bacterial contamination15. We found that the use of staples did not increase the risk of infection when compared with interrupted monofilament sutures. The above findings together support the use of a tourniquet and staples as measures to reduce operative time.
A noteworthy finding of our study was the significantly lower number of infections when a cast was applied in the operating room. This is not surprising because immobilization may have a beneficial effect on soft-tissue recovery. Furthermore, cast application in the operating room may protect the surgical wound from bacterial contamination. It has been recommended that dressings remain unopened for twenty-four to forty-eight hours postoperatively to reduce the infection rate14. The above conclusion seems rational, but the soft-tissue condition may be a confounding factor, since patients with more severe swelling are more likely to have delayed cast application. Although we do not believe that immediate cast application itself prevents deep infection, our findings suggest that a cast should be applied in the operating room provided that soft-tissue injury is minimal.
Properly administered antibiotic prophylaxis significantly reduces the incidence of surgical site infections, and the efficacy of single-dose prophylactic antibiotic therapy has been described15,24-26. To have the desired effect, antibiotic prophylaxis has to be administered within sixty minutes before the incision15. In addition, it has to be fully administered before the tourniquet is inflated14. Olsen et al. demonstrated that suboptimal timing of antibiotic prophylaxis significantly increased the number of postoperative infections following spinal operations16. In the present study, we also noted that suboptimal timing of antibiotic prophylaxis increased the risk of deep surgical site infection following ankle fracture surgery. We observed that, even in our group without infection, antibiotic prophylaxis had been administered suboptimally in 21% of the patients. This is a cause for concern because similar results have recently been reported in another study, reflecting the possible magnitude of this problem16. Suboptimal timing of antibiotic prophylaxis is an important risk factor for infection, and is easily modifiable.
The retrospective study design has inherent limitations. Since deep surgical site infections are relatively uncommon, conducting a prospective study is not feasible. A retrospective age and sex-matched case-control study is an accepted design for an investigation of risk factors for postoperative infection27. However, an important limitation of this study design is the reliance on data provided by the medical and surgical charts. Since patient data are not always properly documented, complications may be underreported. To control for these unavoidable reporting deficiencies, the charts of each patient were scrutinized, and records from all other medical specialties were assessed as well. Another limitation is that some occult infections may not have been identified, and the proportion of deep infections preceded by a superficial infection was not recorded. However, had occult or superficial infections progressed to a deep infection, they would probably have been included in the study population. The strengths of this study include the large number of deep infections identified and treated at a single institution, and an extensive array of evaluated potential risk factors. Furthermore, we believe that the large number of treating surgeons increases the generalizability of the results. Another strength was the inclusion of only deep infections because they can be diagnosed with high specificity and have the greatest impact on clinical outcome. To our knowledge, this is the most comprehensive analysis of risk factors for deep surgical site infection specifically targeting ankle fracture surgery.
Our study revealed several modifiable risk factors predisposing patients to the development of deep surgical site infection following operative treatment of ankle fractures. In the absence of a panacea for postoperative infections, we rely primarily on preventive measures, and therefore recognition of risk factors such as diabetes, smoking, alcohol abuse, and compromised soft tissues is important. Meticulous preoperative planning and implementation of a checklist may be valuable adjuncts in reducing operative time as well as human error. Taken together, the expected impact of these measures on both patient morbidity and health-care costs could be substantial.
SooHoo
NF;
Krenek
L;
Eagan
MJ;
Gurbani
B;
Ko
CY;
Zingmond
DS. Complication rates following open reduction and internal fixation of ankle fractures. J Bone Joint Surg Am.
2009 May;91(
5):1042-9.[CrossRef]
Whitehouse
JD;
Friedman
ND;
Kirkland
KB;
Richardson
WJ;
Sexton
DJ. The impact of surgical-site infections following orthopedic surgery at a community hospital and a university hospital: adverse quality of life, excess length of stay, and extra cost. Infect Control Hosp Epidemiol.
2002 Apr;23(
4):183-9.[CrossRef]
de Lissovoy
G;
Fraeman
K;
Hutchins
V;
Murphy
D;
Song
D;
Vaughn
BB. Surgical site infection: incidence and impact on hospital utilization and treatment costs. Am J Infect Control.
2009 Jun;37(
5):387-97. .[CrossRef]
Wukich
DK;
Lowery
NJ;
McMillen
RL;
Frykberg
RG. Postoperative infection rates in foot and ankle surgery: a comparison of patients with and without diabetes mellitus. J Bone Joint Surg Am.
2010 Feb;92(
2):287-95.[CrossRef]
Schepers
T;
Van Lieshout
EM;
De Vries
MR;
Van der Elst
M. Increased rates of wound complications with locking plates in distal fibular fractures. Injury.
2011 Oct;42(
10):1125-9. .[CrossRef]
Jones
KB;
Maiers-Yelden
KA;
Marsh
JL;
Zimmerman
MB;
Estin
M;
Saltzman
CL. Ankle fractures in patients with diabetes mellitus. J Bone Joint Surg Br.
2005 Apr;87(
4):489-95.[CrossRef]
Costigan
W;
Thordarson
DB;
Debnath
UK. Operative management of ankle fractures in patients with diabetes mellitus. Foot Ankle Int.
2007 Jan;28(
1):32-7.[CrossRef]
Gustilo
RB;
Merkow
RL;
Templeman
D. The management of open fractures. J Bone Joint Surg Am.
1990 Feb;72(
2):299-304.
Pollak
AN;
Jones
AL;
Castillo
RC;
Bosse
MJ;
MacKenzie
EJ; LEAP Study Group. The relationship between time to surgical debridement and incidence of infection after open high-energy lower extremity trauma. J Bone Joint Surg Am.
2010 Jan;92(
1):7-15.[CrossRef]
Nåsell
H;
Ottosson
C;
Törnqvist
H;
Lindé
J;
Ponzer
S. The impact of smoking on complications after operatively treated ankle fractures—a follow-up study of 906 patients. J Orthop Trauma.
2011 Dec;25(
12):748-55.[CrossRef]
Höiness
P;
Engebretsen
L;
Strömsöe
K. Soft tissue problems in ankle fractures treated surgically. A prospective study of 154 consecutive closed ankle fractures. Injury.
2003 Dec;34(
12):928-31.[CrossRef]
Tønnesen
H;
Pedersen
A;
Jensen
MR;
Møller
A;
Madsen
JC. Ankle fractures and alcoholism. The influence of alcoholism on morbidity after malleolar fractures. J Bone Joint Surg Br.
1991 May;73(
3):511-3.
Bachoura
A;
Guitton
TG;
Smith
RM;
Vrahas
MS;
Zurakowski
D;
Ring
D. Infirmity and injury complexity are risk factors for surgical-site infection after operative fracture care. Clin Orthop Relat Res.
2011 Sep;469(
9):2621-30. .[CrossRef]
Bosco
JA
3rd;
Slover
JD;
Haas
JP. Perioperative strategies for decreasing infection: a comprehensive evidence-based approach. J Bone Joint Surg Am.
2010 Jan;92(
1):232-9.
Fletcher
N;
Sofianos
D;
Berkes
MB;
Obremskey
WT. Prevention of perioperative infection. J Bone Joint Surg Am.
2007 Jul;89(
7):1605-18.[CrossRef]
Olsen
MA;
Nepple
JJ;
Riew
KD;
Lenke
LG;
Bridwell
KH;
Mayfield
J;
Fraser
VJ. Risk factors for surgical site infection following orthopaedic spinal operations. J Bone Joint Surg Am.
2008 Jan;90(
1):62-9.[CrossRef]
Wild
S;
Roglic
G;
Green
A;
Sicree
R;
King
H. Global prevalence of diabetes: estimates for the year 2000 and projections for 2030. Diabetes Care.
2004 May;27(
5):1047-53.[CrossRef]
Kao
LS;
Meeks
D;
Moyer
VA;
Lally
KP. Peri-operative glycaemic control regimens for preventing surgical site infections in adults. Cochrane Database Syst Rev.
2009 Jul 8;.
Kean
J. The effects of smoking on the wound healing process. J Wound Care.
2010 Jan;19(
1):5-8.
Høiness
P;
Strømsøe
K. The influence of the timing of surgery on soft tissue complications and hospital stay. A review of 84 closed ankle fractures. Ann Chir Gynaecol.
2000;89(
1):6-9.[PubMed]
Carragee
EJ;
Csongradi
JJ;
Bleck
EE. Early complications in the operative treatment of ankle fractures. Influence of delay before operation. J Bone Joint Surg Br.
1991 Jan;73(
1):79-82.
Smith
TO;
Hing
CB. The efficacy of the tourniquet in foot and ankle surgery? A systematic review and meta-analysis. Foot Ankle Surg.
2010 Mar;16(
1):3-8. .[CrossRef]
Saied
A;
Zyaei
A. Tourniquet use during plating of acute extra-articular tibial fractures: effects on final results of the operation. J Trauma.
2010 Dec;69(
6):E94-7.[CrossRef]
Gillespie
WJ;
Walenkamp
G. Antibiotic prophylaxis for surgery for proximal femoral and other closed long bone fractures. Cochrane Database Syst Rev.
2001;.
Jaeger
M;
Maier
D;
Kern
WV;
Südkamp
NP. Antibiotics in trauma and orthopedic surgery—a primer of evidence-based recommendations. Injury.
2006 May;37
Suppl 2:S74-80.[CrossRef]
Slobogean
GP;
O’Brien
PJ;
Brauer
CA. Single-dose versus multiple-dose antibiotic prophylaxis for the surgical treatment of closed fractures. Acta Orthop.
2010 Apr;81(
2):256-62.[CrossRef]
Busse
JW;
Obremskey
WT. Principles of designing an orthopaedic case-control study. J Bone Joint Surg Am.
2009 May;91
Suppl 3:15-20.[CrossRef]
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.