It is generally accepted that surgeons may achieve more accurate and precise placement of implants using TA-THA. It is unclear whether improved positioning translates into improved clinical outcomes, whether certain subsets of patients benefit from intraoperative technology over conventional techniques, and whether there is improvement in long-term implant survivorship with the use of advanced technology during surgery.[6,9,12,25–27]
Interestingly, TA-THA has not been universally adopted. In the NSQIP data set from 2010 to 2018, only 1.3% of the patients underwent THA with some type of navigation or robotic system. One previous study of the National Inpatient Sample (NIS) database found that the utilization rate of TA-THA gradually increased from 0.1% in 2005 to 3.0% in 2014. Another study of hip and knee arthroplasty procedures in a New York State Payor database found that the utilization of TA-THA increased from 0.5% to 5.2% between 2008 and 2015. Although there is a notable discrepancy between NSQIP and other databases, the overall utilization rate remains relatively low for THA.
Among NSQIP participating hospitals, there were notable differences in the overall racial demographic makeup between the TA-THA and U-THA cohorts. In the TA-THA cohort, White, Hispanic, and Black/African American patients were overrepresented and Asian and Native American patients were underrepresented compared with the U-THA cohort. This indicates that there are some slight, but notable racial differences in the allocation of TA-THA across the US health system. These results differ from those reported from the NIS database, in which Hispanic patients were more likely and Black patients were less likely to undergo TA-THA as compared with White patients. It was noted that patients from western regions of the United States were more likely than those from the northeast to receive TA-THA. Nevertheless, several authors have suggested that socioeconomic factors and patient income markedly affect the likelihood of undergoing TA-THA as well.[16,28,29] Our analysis is somewhat limited in this regard because of the constraints of the NSQIP database, which lacks hospital and geographic data.
To date, this is the largest study to examine differences in surgical time between TA-THA and U-THA. Although previous studies yielded mixed results as to whether TA-THA is associated with increased surgical time,[5,6,20,30–32] we unsurprisingly found that the mean surgical time for TA-THA was markedly greater than that of U-THA by approximately 10 minutes. This result is likely clinically meaningful, especially in the context of high-volume adult reconstruction hip surgeons, and most likely manifests from registering bony landmarks, interpreting intraoperative data, and adjusting implant positions accordingly. However, our findings indicate that TA-THA operators became more efficient during the years of study. The differences in mean surgical times between groups narrowed from 36 minutes in 2010 to 15 minutes in 2018, suggesting that hospital systems along with the surgeons performing TA-THA have become more efficient over time because they emerged from their initial learning curve or that the technology itself has substantially improved and become more efficient over time.
In examining inpatient variables, we found that TA-THA patients had a markedly shorter LOS as compared with U-THA patients. In the existing literature, the association between TA-THA and shorter LOS is unclear. Although several studies support this finding,[6,8,28,33] an NIS database analysis consisting of patients undergoing THA from 2005 to 2014 found no association. In our analysis, patients in the TA-THA cohort were more likely to be discharged home as compared with the U-THA cohort, which is consistent with the prior literature. By performing propensity score matching, our analysis further indicates that patients undergoing surgery using computer assistance benefit from shorter LOS and improved home discharge rates. Notably, institutions financially capable of supporting robotic and/or navigation technology may also have more robust perioperative protocols, same-day surgery programs, and physical therapy resources, possibly leading to improved inpatient outcomes as well.
Although overall major complication rates did not markedly differ between the two groups, TA-THA patients had markedly lower postoperative transfusion rates compared with U-THA patients. Previous studies examining total knee arthroplasty in the NSQIP database recorded similar findings, but no previous study of TA-THA has noted this finding or differences in blood loss between the two groups. In addition, TA-THA had higher postoperative readmission rates as compared with U-THA. Additional analysis revealed that patients in the TA-THA cohort were readmitted at higher rates for gastrointestinal complaints and THA implant complication or dislocation. In manually reviewing the diagnosis codes of readmitted patients, it was difficult to determine whether certain readmissions were due to dislocations or issues with THA implant. Nevertheless, the differences in readmission for medical reasons may be related to a markedly higher number of patients being discharged home in the TA-THA group. Revision surgery and mortality rates did not differ between groups, although patients undergoing TA-THA had a higher rate of unrelated operations in 30 days after THA.
One explanation for the perioperative differences seen between TA-THA and U-THA groups is that providers who perform TA-THA and those who perform U-THA work in different practice environments or different regions of the country. Previous studies pertaining to the utilization of TA-THA have shown that it is more likely to be done at urban and teaching hospitals.[16,28] We postulate that TA-THA cases in the NSQIP database were predominantly done by providers at urban, tertiary referral centers, orthopaedic specialty hospitals, or ambulatory surgery centers. This could provide a possible explanation for the differences in perioperative variables.
Notably, our analysis produced slightly different rates of postoperative complications for the subgroup analysis of major complications, readmissions, and revision surgeries. This discrepancy is because of coding differences within the database for each category of complication. We believe that the infection rates are best approximated in the revision surgery subgroup analysis because the standard of care for a deep implant infection involves surgical management. The NSQIP database likely lacks granularity in ICD diagnosis coding to capture the rates of orthopaedic-specific complications, and therefore, the revision surgery subgroup analysis provides the most accurate complication rates for infection. In addition, complications treated on an outpatient basis or through emergency department care are not recorded, resulting in the underreporting of dislocation events.
This study had several limitations that must be noted. We were unable to detect functional outcomes or radiographic differences between the groups because these data are not recorded in the database. Because this was a retrospective database study, selection bias likely exists between the study groups. We performed propensity score matching to account for potential confounders between patients, but since the data set was completely deidentified, individualized hospital data could not be obtained. These likely influenced our results because there was likely great heterogeneity between hospital systems for demographics, patients' social support, postoperative inpatient care, perioperative rehabilitation protocols, clinical care coordination, and access to multidisciplinary home service. All these factors have been shown to reduce short-term complications and readmission rates.[29,35–37] Because the NSQIP database used CPT codes to record surgical details, we were unable to differentiate between computer navigation and robotic THA. Notably, not all surgeons may include a secondary CPT code for intraoperative technology when submitting procedure codes to the NSQIP database. This would serve to underreport the number of cases that use intraoperative technology within the NSQIP database. Furthermore, NSQIP is limited in the specificity of ICD coding for readmission and revision surgery diagnoses and does not record emergency department visits in which the patient is discharged. This likely caused our study to underrepresent the true complication rates, specifically dislocations where a closed reduction was done successfully in the emergency department. Finally, the NSQIP database only provided a 30-day follow-up, which does not capture the true complication rate within the 90-day bundled period.
In summary, our investigation of TA-THA in the NSQIP database both parallels and differs from that of previously reported demographic differences with TA-THA utilization. This is the largest study to report surgical times for TA-THA. Our analysis revealed that the use of TA-THA adds approximately 10 minutes to mean surgical time compared with U-THA. Notably, this difference narrowed during the study period. Patients who underwent TA-THA were less likely to undergo postoperative transfusion, had a shorter mean LOS, and were more likely to be discharged home. Those undergoing TA-THA did have higher readmission rates as compared with patients undergoing U-THA. These perioperative differences may be explained by differences between providers and hospitals that provide TA-THA as compared with those who only perform U-THA, which is consistent with previous studies that found TA-THA to be done more frequently at urban and teaching hospitals.[16,28] We believe that these results are reflective of this trend. These institutions have a greater capacity for orthopaedic-specific and multidisciplinary home care and are more likely to use protocols that can maximize and improve outcomes. Future research should aim to focus on examining additional factors that influence whether a patient received TA-THA over U-THA, such as hospital environment and geographical region, to assess their effect on patient outcomes and the utility of technology
J Am Acad Orthop Surg. 2022;30(8):e673-e682. © 2022 American Academy of Orthopaedic Surgeons