No Association Found Between the Direct Anterior Approach and Increased PJI RIsk
Experienced surgeons who utilize the direct anterior approach can be confident that they are not putting their patients at increased risk for periprosthetic joint infection (PJI) when they perform total hip arthroplasty (THA) with this approach.
That’s the conclusion of a study from The Rothman Institute presented recently at the annual meeting of the Musculoskeletal Infection Society (MSIS), providing a counterpoint to recent studies that had suggested a link between the direct anterior approach and an increased risk of PJI.
“The purpose of this study was to investigate whether such increased risk does indeed exist, using a large cohort of patients, operated on by experienced surgeons, and taking into account various confounders,” said lead author Noam Shohat, MD, who presented the data during the virtual MSIS meeting.
The researchers, under the direction of senior author Javad Parvizi, MD, identified 9416 consecutive primary THAs performed by 4 Rothman Institute surgeons between 2006 and 2016, including 3959 direct anterior approach THAs and 5457 direct lateral approach THAs. They extracted data on demographics and comorbidities from the patient records and then performed both univariate analysis and multivariate regression to identify and account for various confounders.
Univariate analysis found no association between the surgical approach and PJI, with PJI rates of 0.5% for the direct anterior approach and 0.6% for the direct lateral approach (P=0.265). Dr. Shohat noted that the overall PJI rate in THA patients decreased steadily during the study period, from 2% (20/1005) in 2010 to 0.8% (7/914) in 2018. Surgeons at The Rothman Institute adopted a number of measures, such as the use of dilute povidone-iodine irrigation, skin closure with subcuticular monofilament suture, and application of occlusive dressings, that explain the reduction in the PJI rate over the course of the study period.
In the multivariate regression analysis, the researchers evaluated the effect of several potential risk factors for PJI, including body mass index (BMI), Charlson Comorbidity Index (CCI), operative time, general versus spinal anesthesia, intraoperative povidone-iodine irrigation, monofilament suture closure versus staples, and occlusive dressing use. The direct anterior approach again showed no significant independent association with PJI (adjusted odds ratio 0.81, 95% confidence interval 0.43-1.54; P=0.523).
Dr. Shohat said that their analysis revealed several risk factors that are significantly associated with higher PJI rates, inclulding:
- Male gender (P=0.045)
- Higher BMI levels (P<0.001)
- Higher CCI (P<0.001)
- Higher ASA scores (P<0.001)
- Use of general anesthesia (P<0.001)
- Longer operative time (P<0.001)
- Greater blood loss (P<0.001)
- Shorter length of stay (P<0.001)
Use of dilute povidone-iodine irrigation (P=0.005), monocryl/liquid adhesive closure (P<0.001), and occlusive dressing (P=0.018) were associated with lower rates of PJI, Dr. Shohat said.
The findings of this study contrast with the findings of a study from NYU Langone Health published last year. Aggarwal et al  evaluated 6086 THA patients and found a significantly higher risk for PJI with the direct anterior approach compared with non-direct anterior approach THAs in both univariate and multivariate regression analysis.
Both studies were based on analysis of a large data set from a single institution. Why the diametrically opposed findings?
Dr. Shohat said there are several limitations of the NYU Langone Health study that he and his colleagues sought to address in their study, starting with the length of time between surgery and assessment of PJI. Aggarwal et al  assessed patients for PJI at 3 months after surgery, “which is a relatively short follow-up time and may have affected PJI rates in both direct anterior approach and non-direct anterior approach groups,” Dr. Shohat said. “A 1- or 2-year follow-up is more appropriate in the evaluation of PJI.”
Aggarwal et al  included far fewer confounders in their regression analysis than Dr. Shohat and his colleagues did. “Only BMI and the presence of diabetes mellitus were found to increase the risk for PJI in the univariate analysis [by Aggarwal et al] and, therefore, many potentially important confounders were excluded from multivariate regression analysis,” Dr. Shohat said.
“Factors that have consistently been shown to be associated with PJI, such as length of surgery, gender, and length of stay, were not significant on univariate analysis [in the Aggarwal et al study]. Other important variables, including CCI, ASA, and type of anesthesia, as well as specific changes to PJI prevention practice, were not evaluated. Including these factors in a regression analysis may have significantly altered their results.
“In contrast and in accordance with prior studies, we found multiple factors to be associated with an increased risk for PJI and included them in the regression analysis that showed the approach was not significantly associated with PJI.”
Like the study by Dr. Shohat and colleagues, Aggarwal et al  was conducted at a single institution. But while the study from The Rothman Institute included patient data from only 4 surgeons, the study from NYU Langone Health used patient data from 25 surgeons. “Variations in practice could not be accounted for [in the study by Aggarwal et al],” Dr. Shohat said. “Wide practice variations among a diverse group of surgeons may have influenced the results. In contrast, our focus was on only 4 surgeons and we were able to document the exact changes in practice made throughout the study period.”
The learning curve for the direct anterior approach may also have affected the results of the study by Aggarwal et al,  who were unable to account for the learning curve of each surgeon in the study. “Higher mechanical complication rates and revisions seen in the learning period of the direct anterior approach may have increased PJI rates and biased the conclusion,” Dr. Shohat said. “We limited our analysis to 4 fellowship-trained, high-volume arthroplasty surgeons with substantial experience in the surgical approach used, further reducing the bias from variables not associated with the surgical approach.”
Shohat N, Goswami K, Tan T, Breckenridge L, Gursay D, Clarkson S, Parvizi J. Direct Anterior Approach to the Hip Does not Increase the Risk for Subsequent Periprosthetic Joint Infection. Abstract 25. Presented at the 30th Annual Open Scientific Meeting of the Musculoskeletal Infection Society, August 7-8. The program and abstracts for the meeting can be found here.
- Aggarwal VK, Weintraub S, Klock J, et al. 2019 Frank Stinchfield Award: A comparison of prosthetic joint infection rates between direct anterior and non-anterior approach total hip arthroplasty. Bone Joint J 2019 Jun;101-B(6_Supple_B):2-8. doi: 10.1302/0301-620X.101B6.BJJ-2018-0786.R1.