The last three decades have seen publications on DAA total hip arthroplasty (THA) increase exponentially, with 2019 seeing more than 15 times more PubMed-indexed articles published on the topic than in 1990 (Figure 48.1). Interest in the technique has accelerated in recent years due to the numerous reported advantages over more traditional techniques, many of which have been supported by the expanding research on the DAA. This expansion of literature focused on the DAA has also generated research findings at odds with some of these established advantages, sparking a lively debate within the orthopaedic community over the true benefits and the potential risks of DAA THA.

With high-level literature being published on this topic year over year, keeping abreast of the most pertinent articles and areas of focus is essential for any surgeon regularly performing DAA THAs. In this chapter, applying hierarchal levels to the literature, we review the highest-level DAA literature, focusing in particular on publications from the last 5 years, and summarize the key findings and overall messages. We also appraise research that has produced results that do not favor the DAA and discuss what their findings mean for the future of DAA practice.

Although it is seldom the primary outcome variable in any clinical study, the DAA operative time is one of the most commonly asked questions from patients at the preoperative office visit. Of course, there is a significant amount of room for subjectivity on this topic; studying 100 surgeons’ operative time will yield 100 different results, and many will argue that provided the patient outcome is optimized, operative time is of little consequence. However, with due consideration to health economic factors, it is a useful metric to be able to optimize resources, such as operating room time and the number of cases per surgical list.

Gondusky et al¹ reported the mean operative time per hip as 72 minutes in their study of simultaneous bilateral DAA THA, whereas Matta et al² reported 75 minutes for unilateral DAA hips on an orthopaedic table. Numerous studies have reported on direct comparisons in operative time between the DAA and more traditional approaches. Although Wang et al³ reported no statistically significant differences in operative time when comparing the DAA with the direct lateral (DL) approach, numerous studies have reported longer operative times for the DAA compared with either the posterior approach (PA) or the lateral approach.^4,5,6,7

However, an important factor to consider is where these cases fall in the surgeon’s learning curve. As reported by Stone et al,⁸ DAA cases performed during the learning curve have higher than average operative times recorded, as may be expected during the assimilation of a new operative technique. They also reported that, when the learning curve is completed, operative times for the DAA decrease to within normal parameters compared with other approaches. Virtually none of the studies that reported longer DAA operative times indicated whether the cases included in the study were part of the learning curve or if the senior surgeon performing the procedures had significant experience with the DAA. This may bias the results to artificially longer operative times that are due to inexperience with the approach rather than the approach itself. Longer operative times clearly show that the surgeons performing the DAA were still in their learning curve. Operative times for DAA THA are shorter than other approaches once the surgeon is completely through the learning curve,^8,9 mainly because the closure for the DAA is much easier and faster.

With emerging studies focusing on variations in tranexamic acid protocols¹⁰ and the type of cauterization device used,¹¹ blood loss during DAA surgery is currently an area of keen research interest. Although some recent studies have reported no significant difference in blood loss when comparing the DAA with the anterolateral approach¹² and the PA,¹³ respectively, one study reported greater blood loss in the DAA compared with the PA.¹⁴ Conversely, numerous recent high-level articles have stated that the DAA results in lower blood loss compared with alternative approaches.^15,16,17

Conceptually, one would expect lower levels of bleeding in the DAA because of its muscle-sparing nature. Indeed, aside from encountering the ascending branch of the lateral circumflex vessels during the initial dissection of the approach, no major vascular structure should be encountered at any point during the DAA procedure.

A surrogate outcome related to blood loss that is often reported is the postoperative transfusion rate—an important factor that has been previously linked to higher rates of periprosthetic joint infection. With the advent of tranexamic acid use in total joint arthroplasty, transfusion rates have dramatically decreased irrespective of the surgical approach. However, a study comparing the Watson-Jones approach with the DAA found the decrease in transfusion requirements with the DAA to be statistically significant.¹⁸

Comparing postoperative pain assessments across different research articles can potentially be fraught with inappropriate conclusions; cohort heterogeneity and the use of different pain assessment tools can make external validity difficult to determine. In this area, the recent literature has solidified the DAA’s superiority over alternative approaches.

Much of the latest research documenting postoperative pain score improvements in DAA patients has used the visual analog score as the outcome measure. The quality of the literature runs the gamut of the levels of evidence, including level 1 studies,^{7,15,19,20,21} level 2 studies,²² registry data,²³ and lower-level evidence.²⁴ The DAA advantage is clear; in all patient groups and study types, DAA patients experienced statistically significant decreases in postoperative pain that were superior to the alternative approach comparators. Some studies limited their assessment of this variable to the immediate postoperative period, whereas others measured it into short- and even medium-term follow-ups.

Miller et al¹⁹ reported in their meta-analysis of seven randomized controlled trials and six prospective nonrandomized cohort studies that DAA patients showed improved pain scores and lower narcotic consumption (often used as a surrogate outcome for postoperative pain relief) compared with PA patients. Brismar et al²² found that the statistically significant differences between DAA patient pain scores and DL approach patient scores persisted to 8 weeks postoperatively but disappeared by 1 year. This finding is mirrored in much of the short- to medium-term follow-up literature surrounding the DAA and is often paradoxically used as a criticism of the approach, with “no difference beyond 6 weeks” being a commonly used phrase among DAA skeptics. The wealth of high-level literature on the topic supports the conclusion that DAA patients experience improved pain relief until such time as their pain relief is equivalent to other approaches, thereby proving to be the superior approach overall when examining postoperative pain relief.

Although there are many factors that can influence a patient’s pain after DAA THA, a significant contributing factor is likely the muscle-sparing nature of the approach; less muscular trauma during surgical dissection leads to less pain and a shorter recovery time for the muscles surrounding the surgical field. Several of the studies reporting on postoperative pain relief also report on incision length,^7,15,24 which is consistently found to be shorter in the DAA than with any other approach. Although incision length itself does not likely have a significant effect on postoperative pain, we can infer that a smaller surgical field will likely result in a smaller area of surgical trauma, which could, in turn, affect the patient’s postoperative pain experience. However, this theory is not supported by available evidence.

Although length of stay (or length of hospitalization) is an oft-reported metric and considered a surrogate outcome for the speed of postoperative recovery, it bears consideration that there are multiple factors that will determine how long a patient spends in the hospital after THA using any approach. In North America and throughout the world, variances in postoperative recovery protocols make drawing conclusions a challenge. However, the recent published literature investigating this outcome has yielded interesting findings.

Although a handful of studies report no statistically significant differences in the length of stay between approaches,^7,15 there are a great many more that report clear DAA superiority.^{25,26,27,28,29,30,31} Jia et al²⁵ reported a significantly lower length of stay when comparing DAA patients with PA patients in their level 1 systematic review and meta-analysis including 7377 patients. Zhao et al²⁶ conducted a randomized controlled trial directly comparing the DAA with the posterolateral approach (PLA) in 120 patients and found the mean length of stay for DAA patients was 2.8 days, whereas the mean for PLA patients was 3.3 days (P = .04). These findings were mirrored in a more recent retrospective cohort study by Siljander et al,²⁹ who reviewed 5341 procedures and reported the mean length of stay as 1.8 ± 0.9 days for DAA patients and 2.3 ± 1.4 days for PLA patients.

No recent studies have been published demonstrating a longer length of stay for DAA patients. Martinkovich et al²⁷ recently published a retrospective review of 429 patients who underwent THA via the DAA, reporting that patients whose surgical start times were before 9 AM (effectively translating to the first case of the list) were 11 times more likely to be discharged the same day compared with those with start times after 12 PM. Although not a surprising finding, the authors go on to report that patients who achieved same-day discharge were less likely to return to the emergency room within 90 days, to be readmitted (at both 30 and 90 days), or to require revision surgery. However, it is unclear whether strategic booking of surgical lists biased this outcome.

In facilities that have resources available to support an outpatient THA program, patients are usually screened to establish candidacy; patients with multiple medical comorbidities, those with a history of surgery-related complications (eg, difficulty with pain control, nausea and vomiting), and patients who are deemed too frail are often excluded from the program, introducing selection bias. They are also often booked as the first case in the morning to maximize their recovery time in the hospital and thus the probability that they will not require an inpatient bed overnight. Indeed, the article goes on to recommend that patients selected for same-day discharge should receive earlier operating room start times, effectively acknowledging the bias of their study.

One of the most frequently quoted benefits to performing DAA hip arthroplasty over any of the other approaches is that DAA patients tend to recover quicker within the early postoperative period. There is ample literature to support this claim,^26,28,32,33 and recent publications on the topic have generally supported the DAA in this outcome measure.