Abstract
Background
Numerous studies have shown that elevated BMI is associated with adverse outcomes in THA; however, BMI alone does not adequately represent a patient’s adipose and soft tissue distribution, especially when the direct-anterior approach is evaluated. Local soft tissue and adipose, especially in the peri-incisional region, has an unknown impact on patient outcomes after direct-anterior THA. Moreover, there is currently no known evaluation method to estimate the quantity of local soft tissue and adipose tissue. The current study introduced a new radiographic parameter that is measurable on supine AP radiographs: the abdominal pannus sign.
Question/purpose
Are patients who have an abdominal pannus extending below the upper (cephalad) border of the symphysis pubis more likely to experience problems after anterior-approach THA that are plausibly associated with that finding, including infections resulting in readmission, wound complications resulting in readmission, fractures, or longer surgical time, than patients who do not demonstrate this radiographic sign?
Methods
Between 2015 and 2020, five surgeons performed 727 primary direct-anterior THAs. After exclusion criteria were applied, 596 procedures were included. Of those, we obtained postoperative radiographs in the postanesthesia care unit in 100% of procedures (596 of 596), and 100% of radiographs (596) were adequate for review in this retrospective study. The level of the pannus in relation to the pubic symphysis was assessed on immediate supine postoperative AP radiographs of the pelvis: above (pannus sign 1), between the upper and lower borders (pannus sign 2), or below the level of the pubic symphysis (pannus sign 3). In this study, we combined pannus signs 2 and 3 into a single group for analysis not only because there was a limited number of patients in each group, but also because there was no statistically significant difference between the two groups. Pannus sign 1 was identified in 82% of procedures (486 of 596), and pannus sign ≥ 2 was identified in 18% (110). We compared the groups (pannus sign 1 versus pannus sign ≥ 2) in terms of the percentage of patients who experienced problems within 90 days of THA that might be associated with that physical finding, including infections resulting in readmission including subcutaneous, subfascial, and prosthetic joint infections; wound complications resulting in readmission, defined as dehiscence or delayed healing; and all fractures, and we compared the groups in terms of surgical time—that is, the cut-to-close time.
Results
Patients with a pannus sign of ≥ 2 were more likely than those with a pannus sign of 1 to have a postoperative infection (6.4% [seven of 110 procedures] versus 0.6% [three of 486], odds ratio 10.96 [95% confidence interval (CI) 2.83 to 42.38]; p < 0.01), wound complications (0.9% [one of 110] versus 0% [0 of 486] with an infinite odds ratio [95% CI indeterminate]; p = 0.18), and fractures (4.5% [five of 110] versus 0% [0 of 486], with an infinite odds ratio [95% CI indeterminate]; p < 0.01). The mean surgical time was longer in patients with a pannus sign of ≥ 2 than it was in those with a pannus sign of 1 (128 ± 25.3 minutes versus 118 ± 27.5 minutes, mean difference 10 minutes; p < 0.01).
Conclusion
Based on these findings, patients who have an abdominal pannus that extends below the upper (cephalad) edge of the pubic symphysis are at an increased risk of experiencing serious surgical complications. If THA is planned in these patients, an approach other than the direct-anterior approach should be considered. Surgeons performing THA who do not obtain supine radiographs preoperatively should use a physical examination to evaluate for this finding, and if it is present, they should use an approach other than the direct-anterior approach to minimize the risk of these complications. Future studies might compare the abdominal pannus sign using standing radiographs, which are used more often, with other well-documented associated risk factors such as elevated BMI or higher American Society of Anesthesiologists classification.
Level of Evidence
Level III, retrospective cohort study.
Introduction
THA through the direct-anterior approach has been growing in popularity. This approach has several advantages: early functional recovery, less postoperative pain, earlier cessation of walking aids, and improved acetabular cup placement [10, 18]. However, complications are also a concern. These are frequently related to obesity, with several studies showing an increased incidence of complications in patients with obesity such as wound dehiscence, increased infection risk, and early fracture [4, 12, 22].
The WHO defines obesity as a BMI > 30 kg/m2 [6]. Rates have nearly tripled since the 1960s in the United States, with projections to increase by 33% during the next 20 years [7]. Obesity as it relates to hip and knee arthroplasty is a known risk factor for poor outcomes, particularly periprosthetic joint infection [3, 9, 10, 13, 22]. Identifying correctable risk factors is essential in order to reduce complications [22]. Although there is a clear association between obesity and an increased risk of surgical complications, the most common measure of obesity, BMI, has notable shortcomings. It does not accurately represent the distribution of soft tissue, nor can it differentiate muscle from fat. Previous studies have investigated local adipose measurement and its correlation to infection [9, 14, 17, 23, 24]. However, many of these studies have been centered around soft tissue thickness in the knee and its relation to periprosthetic joint infections [5].
To our knowledge, there have been no studies evaluating the association of local soft tissue on postoperative complications in THA, specifically through the anterior approach. Here, we introduce the pannus sign, which considers the distal-most extent of a patient's pannus in relation to the pubic symphysis on supine AP radiographs of the pelvis. We asked: Are patients who have an abdominal pannus extending below the upper (cephalad) border of the symphysis pubis more likely to experience problems after anterior-approach THA that are plausibly associated with that finding, including infections resulting in readmission, wound complications resulting in readmission, fractures, or longer surgical time, than patients who do not demonstrate this radiographic sign?
Patients and Methods
Study Design and Setting
This was a retrospective, comparative study using our institution’s electronic medical record. Our institution is a community hospital in a coastal community with a large catchment area serving approximately 850,000 people. The practitioners included in this study were subspecialist orthopaedic surgeons and general orthopaedists.
Patients
We used our institution's electronic medical record to access patient data and radiographs for all direct-anterior THAs that were performed from January 2015 to June 2020. Five surgeons at a single institution were included. The cohort of surgeons included two fellowship-trained arthroplasty surgeons (JH and FK), one fellowship-trained sports medicine surgeon (PF), and two general orthopaedists (TG and LS), only one of whom is an author on this study.
Between January 1, 2015, and June 23, 2020, we performed 727 anterior-approach THAs in 658 patients. After excluding 131 procedures performed for femoral neck fractures (75), patients without adequate radiographs (22), bilateral procedures performed on the same day (18), and revision procedures (16), we had 570 patients with 596 procedures. All patients had supine AP radiographs that were accessible for analysis (Fig. 1).
Fig. 1.
This patient selection flowchart illustrates the original data query with refinement using exclusion criteria.
Participants’ Baseline Data
A total of 596 procedures were included in the study. Pannus signs 2 and 3 (defined in the next section) were combined because not enough patients comprised these two groups for statistical analysis. A pannus sign 1 was identified in 82% of surgeries (486 of 596), and a pannus sign of greater than 1 was identified in 18% (110 of 596). The pannus sign 1 group had older patients (mean of 69 ± 11 years versus 65 ± 10 years; p < 0.001) and a higher percentage of women (p = 0.002). There was no difference in the proportion of patients with diabetes between the groups (Table 1).
Table 1.
Patient demographics by pannus sign (n = 596 procedures)
| Parameter | Pannus 1 (n = 486 procedures) | Pannus ≥ 2 (n = 110 procedures) | p value |
| % men | 43 (212) | 59 (65) | 0.002 |
| % with diabetes | 7 (36) | 8 (9) | 0.67 |
| Age in years | 69 ± 11 | 65 ± 10 | < 0.001 |
| Patient height in cm | 170 ± 11 | 168 ± 10 | 0.07 |
| Patient weight in kg | 80 ± 18 | 94 ± 20 | < 0.01 |
| Patient BMI in kg/m2 | 27 ± 5 | 33 ± 5 | < 0.01 |
Data provide as % (n) or mean ± SD. Pannus sign 1: pannus above the superior border of the pubic symphysis; pannus sign ≥ 2: pannus below the superior border of the pubic symphysis.
Defining the Pannus Sign
We used supine AP pelvis radiographs obtained in the postanesthesia care unit for our analysis. This was the most expedient and efficient way to obtain uniform radiographs from different surgeons in a centralized PACS system at our institution. Ideally, the radiography evaluation would occur preoperatively and with the patient in the supine position to demonstrate the pannus position during hip arthroplasty. Radiographs were used to test a potentially novel technique for correlating outcomes with a pannus sign, as opposed to a traditional physical examination, which should be done in conjunction with radiographs.
The radiographs were taken with the patient lying completely flat on a gurney. An acceptable radiograph met the following criteria: the center point of the radiograph was the pubic symphysis; the collimation included the skin margins laterally, above the iliac crests superiorly, and distal to the femoral stem component inferiorly in a landscape orientation; the obturator foramina appeared symmetrical; the iliac wings had equal concavity; and the greater trochanters of the proximal femur were in profile. We then found the most-distal extent of the abdominal pannus and drew a horizontal line across the bottom of it. The position where this line intersects the patient’s pelvis was evaluated, and each patient was filtered into one of three groups: above the superior border of the pubic symphysis (pannus sign 1), between the upper and lower border of the pubic symphysis (pannus sign 2), and below the lower border of the pubic symphysis (pannus sign 3) (Fig. 2). If the patient did not have a pannus, they were given a Group 1 designation.
Fig. 2.
These AP radiographs of the pelvis demonstrate (A and B) a pannus sign 1, above the level of the pubic symphysis; (C) a pannus sign 2, between the upper and lower border of the pubic symphysis; and (D) a pannus sign 3, below the lower border of the pubic symphysis.
Measurements were made by two blinded orthopaedic surgery residents (AD and NB) who did not have access to any patient information, including BMI. If there was discrepancy, a third blinded orthopaedic surgeon (BS) made the final decision. Interobserver agreement was high at 94% (kappa 0.814 ± 0.073 [95% CI 0.67 to 0.96). Pannus sign 2 and pannus sign 3 designations were combined in the analysis because of the limited number of patients in each category, without any meaningful differences between the two. When combined with one another and then compared with pannus sign 1, a statistically significant difference could be found.
Primary and Secondary Study Outcomes
Our primary study goal was to develop a useful radiographic tool to approximate an abdominal pannus before attempting THA with a direct-anterior approach. This was done by orthopaedic surgeons who had high interobserver reliability when reading AP pelvis radiographs.
Our secondary study goals were to correlate a higher pannus sign (pannus sign ≥ 2) with adverse outcomes within 90 days of THA that resulted in readmission including infection (defined as subcutaneous, subfascial, or prosthetic joint infection), wound dehiscence, and fractures. We also reviewed inpatient data including length of the procedure from skin incision to closure.
Ethical Approval
Our institutional review board approved this study.
Statistical Analysis
An alpha level of 0.05 was used to determine statistical differences. Continuous variables are expressed as means ± standard deviations, and categorical data are presented as percentages. Using the cutoff for a pannus sign of 1, we grouped patients into risk groups. Demographic characteristics were compared between patients in the pannus sign 1 group and patients with a pannus sign greater than 1 and between patients who were readmitted and patients who were not (Table 2). The analysis was performed using Fisher exact tests to determine odds ratios, and a p value of < 0.05 was considered to indicate a significant difference between the different pannus types. The statistical analysis was performed using R Studio Desktop, Version 1.4.1103.
Table 2.
Reasons for readmission by pannus sign (n = 596 procedures)
| Reason for readmission | Pannus sign ≥ 2, % (n) (n = 110 procedures) | Pannus sign 1, % (n) (n = 486 procedures) | OR (95% CI) | p value |
| Fracture | 4.5 (5) | 0 (0) | Infinite (indeterminate) | 0.005 |
| Infection | 6.4 (7) | 0.6 (3) | 10.96 (2.83 to 42.38) | 0.008 |
| Wound dehiscence | 0.9 (1) | 0 (0) | Infinite (indeterminate) | 0.18 |
Pannus sign 1: pannus above the superior border of the pubic symphysis; pannus sign ≥ 2: pannus below the superior border of the pubic symphysis.
Results
Complications and Surgical Time
Patients with a pannus sign of ≥ 2 were more likely than those with a pannus sign of 1 to have a postoperative infection (6.4% [seven of 110 procedures] versus 0.6% [three of 486], OR 10.96 [95% confidence interval (CI) 2.83 to 42.38]; p < 0.01), wound complications (0.9% [one of 110] versus 0% [0 of 486], odds ratio infinity [95% CI indeterminate]; p = 0.18), and fractures (4.5% [five of 110] versus 0% [0 of 486], odds ratio infinity [95% CI indeterminate ]; p < 0.01). The mean surgical time was longer in patients with a pannus sign of ≥ 2 than it was in those with a pannus sign of 1 (128 ± 25.3 minutes versus 118 ± 27.5 minutes, mean difference 10 minutes; p = 0.002). There were 37 readmissions within 90 days postoperatively. Nineteen of the 37 readmissions were in the 110 procedures in patients with a pannus sign of 2 or 3, or 17% (19 of 110), while 18 were in the 486 procedures in patients with a pannus sign of 1, or 4% (18 of 486). A total of 51% (19 of 37) of these readmissions were in patients with a pannus sign of 2 or 3, whereas procedures in patients with a pannus sign 1 comprised 49% (18 of 37). The reasons for readmission, including fracture, infection, and wound dehiscence, were not dissimilar between groups (Table 2). Those who were readmitted for any reason had an increased time from incision to closure (133.8 minutes versus 118.7 minutes; p = 0.04) as well as hospital length of stay (3 days versus 1.9 days; p < 0.01) compared with patients who were not readmitted.
Discussion
BMI alone does not adequately represent a patient’s adipose and soft tissue distribution, especially when evaluating the direct-anterior approach in THA. It ignores important factors such as muscle distribution and local soft tissue or adipose tissue, especially in the peri-incisional region. Until now, the pannus level had an unknown impact on patient outcomes after direct-anterior THA. Our study found that a large pannus that drapes below the level of the upper border of the pubic symphysis, as seen on postoperative AP pelvis radiographs, is associated with a longer hospital stay and duration of surgery, as well as an increased odds of fracture and infection resulting in readmission within 90 days. Based on our findings, surgeons have data to support recommending an approach other than the direct-anterior for patients with a pannus sign of 2 or greater.
Limitations
There are several limitations to our study. First a selection bias may have been influenced by several patients who could have been excluded from this study because of unusually high BMIs, pannus sign notwithstanding. In practice, our institution typically uses a BMI cutoff of 35 kg/m2, possibly excluding many patients with higher pannus signs. Further, although we attempted to standardize postoperative radiographs, radiographs often appeared dissimilar in quality. This, however, was not a disqualifying problem because any radiographs that did not clearly display the skin margins and appropriate bony anatomic landmarks were not included in our study.
The biggest limitation of our study is sparse data. Despite finding statistically significant differences in our endpoints, our ORs and associated CIs were large or even infinite or undefined. This is a major drawback. However, although we cannot conclude causation without a doubt from our findings, we present a trend toward higher adverse event rates in patients with higher pannus signs with a reproducible diagnostic examination finding, providing practitioners with datapoints with which they can make decisions about surgical approaches. This limitation may also be ameliorated with more datapoints and a larger sample of patients. Another limitation is the subjective nature of several endpoints, particularly readmission. Emergency medicine providers typically decide who to admit to the hospital. Although providers may defer to admission thresholds, the reasons for admissions were not dissimilar.
Another limitation in our study is that we did not compare the pannus sign with BMI or perform a multivariate analysis based on sex. BMI may be a major confounder in that those with higher BMIs are likely to also have larger abdominal panni. The exception to this is patients who are more muscular or those who are shorter, in whom BMI may be artificially elevated. In the present study, we avoided this comparison because we felt that with the limited number of patients in our study, we would not be able to perform an adequate univariate analysis. In a future study with a larger sample size, we might be able to more thoroughly compare BMI with the pannus sign and rule it out as a confounding variable in addition to performing a sex-based multivariate analysis.
Radiographs were taken postoperatively in the postanesthesia care unit, which is a limitation. To be a useful indicator, radiographs should be taken preoperatively. Radiographs were used in this manner because of the ease of obtaining a large dataset from the unified PACS system at our institution. This study serves to help validate a diagnostic test with high interobserver reliability to identify a potential risk factor for direct-anterior THA. Further, we thought it useful to identify where the pannus may lie during the surgical approach. In the future, we might compare supine radiographs with preoperative standing radiographs, which are ubiquitous in arthroplasty clinics. In addition, we compared standing radiographic pannus signs with physical examination findings. Although a physical examination may be a cheap method of understanding the pannus’s location and adipose distribution, it may not be as accurate or quantifiable as radiographs, especially with bony anatomy that is difficult to palpate in patients with obesity.
Some of the patients with surgical site infections or wound problems were treated as outpatients with local wound care or oral antibiotics, in which case our study would have missed this patient population, given that we only evaluated 90-day hospital readmissions. We estimate that the actual number of postoperative complications is higher than our study indicates. However, there is no reason to believe this would be more likely in one group than the other.
Discussion of Key Findings
Primarily, our study’s goal was to develop a useful radiographic tool to approximate an abdominal pannus, which was done with high interobserver reliability. Although a physical examination could suffice in understanding local soft tissue and adipose distribution, it may be inaccurate in people with obesity; radiography can be a useful, relatively inexpensive adjunct to translate this information across practice environments and to perform future studies with, correlating this diagnostic test with outcomes in larger patient populations and with multivariate analyses. From these data, orthopaedic surgeons performing direct-anterior THA may decide to alter surgical approaches in patients with a pannus sign of 2 or greater as seen on AP pelvis radiographs—radiographs they would otherwise be obtaining preoperatively, albeit while the patient is standing.
Our secondary study goals were to correlate a higher pannus sign with adverse outcomes within 90 days of THA that resulted in readmission. Patients with an abdominal pannus seen to overhang the symphysis pubis on a supine radiograph were more likely to experience complications after THA performed through an anterior approach, including infection, wound complications, and fractures. They also had a longer mean surgical time and higher rate of all-cause readmission. These data are supported elsewhere regarding THA and TKA. Purcell et al. [20] reported that a BMI greater than 35 kg/m2 is associated with increased postoperative infection rates that result in revision after direct-anterior THA. They postulated this was because of the proximity of the incision to the genitalia, inguinal crease, and overlying abdominal pannus. In a separate study, Purcell et al. [19] found that patients with anterior hip procedures experienced more superficial wound complications than those with the posterior approach across all BMIs. Regarding periprosthetic joint infection after TKA, studies have used soft tissue thickness as a marker [22, 24]. Soft tissue thickness was measured on lateral radiographs of the knee. These studies concluded that patients who had greater soft tissue thickness had a higher risk of periprosthetic joint infection after TKA, independent of BMI.
With regard to future work, studies might perform multivariate analyses and use radiographs taken with the patient in the standing position as opposed to the supine position. A head-to-head comparison of direct-anterior and posterior approaches regarding the pannus sign may also serve to support our conclusions that an elevated pannus sign is associated with not only adverse events, but also the direct-anterior approach.
In terms of translating our work into practice, we recommend physicians develop new risk stratification strategies before deciding to perform surgery. Rather than simply evaluating BMI, surgeons may be able to take an objective measurement tool, such as an AP pelvis radiograph, and visualize where the abdominal pannus may lie relative to the hip. If the direct-anterior approach is the only approach a surgeon is comfortable with, then recommendations may be made regarding the pannus. Bariatric surgery before joint arthroplasty has become an area of interest to help mitigate the risks associated with obesity and arthroplasty. Although it has not been shown to directly correlate with decreased arthroplasty complications, it has been shown to help with tighter glucose control and improved cholesterol while decreasing BMI, and may aid in managing the size of the pannus, with or without panniculectomy [16, 19]. Further, surgeons may use protective strategies against wound complications and infection in patients with a higher pannus sign. For example, although there are studies to support drains being effective at preventing surgical site infection [2, 8, 14, 15, 18, 20], others have found these drains do not reduce the incidence of infection, but these studies were not limited to high-risk patients. Other options such as negative pressure wound therapy or silver-impregnated dressings, both of which are durable protective options, can be used [1, 11, 21, 25].
Conclusion
THA through the direct-anterior approach has been shown to have unique complications. Our study introduced the pannus sign, which gives surgeons an efficient radiographic assessment tool to help with patient selection. It can also help surgeons counsel patients undergoing direct-anterior THA about their increased risk of periprosthetic fracture and infection, as well as longer operative time and hospital stay. Future studies might aim to expand the sample size of patients, use standing radiographs, and compare the pannus sign with BMI with respect to surgical approach.
Acknowledgments
We thank Dr. Thomas Golden (TG), Dr. Larry Schatz (LS), Dr. Farzin Kabaei (FK), Dr. Petros Frousiakis (PF), and Dr. Jason Hofer (JH) for allowing us to use their surgical data for this study.
Footnotes
Each author certifies that there are no funding or commercial associations (consultancies, stock ownership, equity interest, patent/licensing arrangements, etc.) that might pose a conflict of interest in connection with the submitted article related to the author or any immediate family members.
All ICMJE Conflict of Interest Forms for authors and Clinical Orthopaedics and Related Research® editors and board members are on file with the publication and can be viewed on request.
Ethical approval for this study was obtained from Community Memorial Health System, Ventura, CA, USA (number 020-HSR001).
Contributor Information
Atul Saini, Email: asaini@cmhshealth.org.
Nathan Bascom, Email: nbascom@cmhshealth.org.
Brent Sanderson, Email: bsanderson@cmhshealth.org.
Thomas Golden, Email: tgolden@cmhshealth.org.
References
- 1.Ailaney Nikhil, Johns WL, Golladay GJ, Strong B, Kaloe NV. Closed incision negative pressure wound therapy for elective hip and knee arthroplasty: a systematic review and meta-analysis of randomized controlled trials. J Arthroplasty. 2021;36:2402-2411. [DOI] [PubMed] [Google Scholar]
- 2.Baier PK, Gluck NC, Baumgartner U, Adam U, Fischer A, Hopt UT. Subcutaneous Redon drains do not reduce the incidence of surgical site infections after laparotomy. A randomized controlled trial on 200 patients. Int J Colorectal Dis. 2010;25:639-643. [DOI] [PubMed] [Google Scholar]
- 3.Chesney D, Sales J, Elton R, Brenkel IJ. Infection after knee arthroplasty a prospective study of 1509 cases. J Arthroplasty. 2008;23:355-359. [DOI] [PubMed] [Google Scholar]
- 4.Christensen CP, Karthikeyan T, Jacobs CA. Greater prevalence of wound complications requiring reoperation with direct anterior approach total hip. J Arthroplasty. 2014;29:1839-1841. [DOI] [PubMed] [Google Scholar]
- 5.Electricwala AJ, Jethanandani RG, Narkbunnam R, et al. Elevated body mass index is associated with early total knee revision for infection. J Arthroplasty. 2017;32:252-255. [DOI] [PubMed] [Google Scholar]
- 6.Fehring TK, Odum SM, Griffin WL, Mason JB, McCoy TH. The obesity epidemic: its effect on total joint arthroplasty. J Arthroplasty. 2007;22:71-76. [DOI] [PubMed] [Google Scholar]
- 7.Finkelstein EA, Khavjou OA, Thompson H, et al. Obesity and severe obesity forecasts through 2030. Am J Prev Med. 2012;42:563-570. [DOI] [PubMed] [Google Scholar]
- 8.Fujii T, Tabe Y, Yajima R, et al. Effects of subcutaneous drain for the prevention of incisional SSI in high-risk patients undergoing colorectal surgery. Int J Colorectal Dis. 2011;26:1151-1155. [DOI] [PubMed] [Google Scholar]
- 9.Fujii T, Tsutsumi S, Matsumoto A, et al. Thickness of subcutaneous fat as a strong risk factor for wound infections in elective colorectal surgery: impact of prediction using preoperative CT. Dig Surg. 2010;27:331-335. [DOI] [PubMed] [Google Scholar]
- 10.Hamilton WG, Parks NL, Huynh C. Comparison of cup alignment, jump distance, and complications in consecutive series of anterior approach and posterior approach total hip arthroplasty. J Arthroplasty. 2015;30:1959-1962. [DOI] [PubMed] [Google Scholar]
- 11.Howell RD, Hadley S, Strauss E, Pelham FR. Blister formation with negative pressure dressings after total knee arthroplasty. Curr Orthop Pract. 2011;22:176-179. [Google Scholar]
- 12.Jewett BA, Collis DK. High complication rate with anterior total hip arthroplasties on a fracture table. Clin Orthop Relat Res. 2011;469:503-507. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kozlow JH, Lisiecki J, Terjimanian MN, et al. Cross-sectional area of the abdomen predicts complication incidence in patients undergoing sternal reconstruction. J Surg Res. 2014;192:670-677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee JJ, Odeh KI, Holcombe SA, et al. Fat thickness as a risk factor for infection in lumbar spine surgery. Orthopedics. 2016;39:e1124-e1128. [DOI] [PubMed] [Google Scholar]
- 15.Magann EF, Chauhan SP, Rodts-Palenik S, Bufkin L, Martin JN, Jr, Morrison JC. Subcutaneous stitch closure versus subcutaneous drain to prevent wound disruption after cesarean delivery: a randomized clinical trial. Am J Obstet Gynecol. 2002;186:1119-1123. [DOI] [PubMed] [Google Scholar]
- 16.Martin JR, Watts CD, Taunton MJ. Bariatric surgery does not improve outcomes in patients undergoing primary total knee arthroplasty. Bone Joint J. 2015;97:1501-1505. [DOI] [PubMed] [Google Scholar]
- 17.Mehta AI, Babu R, Karikari IO, et al. 2012. Young Investigator Award winner: the distribution of body mass as a significant risk factor for lumbar spinal fusion postoperative infections. Spine (Phila Pa 1976) . 2012;37:1652-1656. [DOI] [PubMed] [Google Scholar]
- 18.Nakata K, Nishikawa M, Yamamoto K, Hirota S, Yoshikawa H. A clinical comparative study of the direct anterior with mini-posterior approach: two consecutive series. J Arthroplasty. 2009;24:698-704. [DOI] [PubMed] [Google Scholar]
- 19.Purcell RL, Parks NL, Cody JP, Hamilton WG. Comparison of wound complications and deep infections with direct anterior and posterior approaches in obese hip arthroplasty patients. J Arthroplasty. 2018;33:220-223. [DOI] [PubMed] [Google Scholar]
- 20.Purcell RL, Parks NL, Gargiulo JM, Hamilton WG. Severely obese patients have a higher risk of infection after direct anterior approach total hip arthroplasty. J Arthroplasty. 2016;31:162-165. [DOI] [PubMed] [Google Scholar]
- 21.Tyagi V, Kahan J, Huang P, Leslie MP, Rubin LE, Gibson DH. Negative pressure incisional therapy and infection after direct anterior approach primary total hip arthroplasty. Orthopedics. 2019;42:e539-e544. [DOI] [PubMed] [Google Scholar]
- 22.Vahedi H, Ward DT, Lee YS, Shohat N, Chen AF. Greater knee soft tissue thickness predisposes patients to subsequent periprosthetic joint infection after total knee arthroplasty. J Arthroplasty. 2020;35:1924-1927. [DOI] [PubMed] [Google Scholar]
- 23.Wagner ER, Kamath AF, Fruth KM, Harmsen WS, Berry DJ. Effect of body mass index on complications and reoperations after total hip arthroplasty. J Bone Joint Surg Am. 2016;98:169-179. [DOI] [PubMed] [Google Scholar]
- 24.Watts CD, Houdek MT, Wagner ER, Taunton MJ. Subcutaneous fat thickness is associated with early reoperation and infection after total knee arthroplasty in morbidly obese patients. J Arthroplasty. 2016;31:1788-1791. [DOI] [PubMed] [Google Scholar]
- 25.Newman JM, Siqueira MBP, Klika AK, Molloy RM, Barsoum WK, Higuera CA. Use of closed incisional negative pressure wound therapy after revision total hip and knee arthroplasty in patients at high risk for infection: a prospective, randomized clinical trial. J Arthroplasty. 2019;34:554-559. [DOI] [PubMed] [Google Scholar]