Total hip arthroplasty (THA) is an effective treatment for patients with debilitating hip pathology.1 However, complications such as dislocation, infection, and persistent pain can be detrimental to outcomes and have been shown to occur at higher rates in patients who have concomitant hip and spine pathology.2,3 Furthermore, several studies evaluating the surgical treatment of spine and hip pathology have shown that THA patients with previous lumbar spinal fusion (LSF) are at increased risk of dislocation and revision.4–6 Indeed, the spine–hip relation has been frequently implicated to directly influence the risk of prosthetic impingement and dislocation.2,7
Currently, there is a paucity of literature investigating the surgical sequence of THA and LSF and how sequence may affect outcomes. In addition, despite the growing concern of opioid use within the United States, little is known on how the sequence of these procedures influences postoperative pain and opioid use.
The purpose of this study was to determine the impact LSF has on the outcomes of THA. We first sought to investigate the risk of dislocation, revision surgery, and prolonged opioid use in patients who have a remote history of LSF prior to THA. We hypothesized that THA patients with remote history of LSF would have higher rates of complications than THA patients with no history of LSF. Secondarily, we assessed timing of LSF relative THA to determine how operative sequence would affect THA outcomes. We hypothesized that patients who underwent LSF prior to THA are at an increased risk of complications, revisions, and prolonged opioid use following their THA compared with those patients who undergo LSF following THA.
MATERIALS AND METHODS
The PearlDiver Patient Records Database (www.pearldiverinc.com) was utilized for this study. PearlDiver is a national database containing deidentified Humana Inc and Medicare medical records capturing around 25 million records between 2007 and 2017.
Patients who underwent THA between 2007 and first quarter of 2017 were identified with first instance Current Procedure Terminology (CPT) codes (Table 1).
Patients with history of LSF were also identified by CPT code (Table 1). Presence of lumbar spine pathology was identified through International Classification of Disease, Ninth Revision (ICD-9-CM) and Tenth Revision (ICD-10-CM) codes (Appendix A, http://links.lww.com/BRS/B497). With these populations, patients were divided into four cohorts: 1) Primary THA without history of LSF, 2) primary THA that had remote history of LSF (at least 2 yrs prior) prior to hip pathology, and THA patients with concurrent presence of hip and spinal pathology that underwent, 3) THA following LSF, and 4) THA prior to LSF (Figure 1).
ICD-9-CM and ICD-10-CM codes were used to identify postoperative complications within 30 days of surgery, including prosthetic hip dislocation, mechanical/osteolytic complications, prosthetic joint infection, surgical site complication, deep vein thrombosis, neurologic complications, respiratory complications, cardiac complications, death, and an aggregate total of the complications (Appendix B, http://links.lww.com/BRS/B498). Revision THA was specified by CPT code as well (Table 1). Preoperative opioid utilization was determined based on records of a filled opioid prescription within 3 months of index surgery.8 Postoperative opioid utilization was assessed at 1, 3, 6, and 12-month time points with use for greater than 3 months considered to be prolonged.9–11
Descriptive statistics were generated for each patient cohort. Bivariate analysis of complication occurrence was conducted using chi-square analysis. Odds of complication were computed directly as the number of events occurring within 30 days of total hip replacement. Odds of opioid use in a specific month following surgery were also evaluated. Multivariable logistic regression was used to calculate odds ratios. All regression modeling controlled for age, sex, race, and Charlson Comorbidity Index. Logistic regression of opioid utilization also controlled for preoperative opioid use. Statistical analysis was performed using the PearlDiver software, built on R, Version 1.1.442 (RStudio Inc, Boston, MA). An α value of 0.05 was set as the level of significance.
Between 2007 and the first quarter of 2017, a total of 85,595 patients underwent primary THA and were included in this study. Of these patients, 80,131 (93.6%) underwent THA alone and 621 (0.7%) had a remote history of LSF. For patients with concurrent hip and spine pathology, 2016 (2.4%) patients underwent THA following LSF and 1356 (1.6%) underwent THA prior to LSF (Appendix C, http://links.lww.com/BRS/B499). The median time for patients who had LSF following THA was 612.5 days. For patients who had LSF prior to THA, the median time between procedures was 516.5 days.
Complications, Revisions, and Prolonged Opioid Use in Patients With Remote History of LSF or THA Only
THA patients with remote history of LSF prior to hip pathology diagnosis and THA did not experience increased risk of total 30-day complications compared with patients receiving THA only (9.0% (n = 51) versus 7.9% (n = 5826), P = 1.000). Following multivariable analysis, THA patients with remote history of LSF did not face increased risk of total 30-day complications (OR = 1.00, 95% CI: 0.74–1.32, P = 0.9305) or dislocation (aOR 1.28, 95% CI: 0.80–2.05, P = 0.3060) (Table 2). However, THA patients with remote history of LSF experienced increased risk of revision (aOR 1.89, 95% CI 1.38–2.51, P < 0.0001) (Table 2). Opioid utilization of THA patients with prior LSF was greater than THA patients with no history of LSF at all studied postoperative time points (Figure 2).
THA patients with prior LSF were also more likely to fill opioid prescriptions at 1 month (aOR 1.44, 95% CI: 1.19–1.74, P = 0.0002), 3 months (aOR 1.46, 95% CI: 1.18–1.81, P = 0.0004), 6 months (aOR 1.53, 95% CI 1.22–1.90, P = 0.0002), and 12 months (aOR 1.56, 95% CI: 1.23–1.97, P = 0.0002), even after adjusting for preoperative use (Table 2).
Complications, Revisions, and Prolonged Opioid Use in Patients With Diagnosed Spinal Pathology Undergoing LSF Before or After THA
In multivariable analysis to evaluate surgical sequence, we found that dislocation risk was greater for patients who underwent THA prior to LSF (aOR: 2.46, 95% CI: 1.90–3.12, P < 0.0001) as compared with those who had THA following LSF (aOR 1.29, 95% CI: 0.99–1.65, P = 0.0529) (Table 3). Similarly, patients who had THA prior to LSF had higher rates of prosthetic joint infection (aOR: 2.65, P < 0.0001 vs. aOR: 1.03, P = 0.8160) and surgical site complications (aOR 1.53, P = 0.0266 vs. aOR 1.46, P = 0.5194) compared with those undergoing THA following LSF.
Patients undergoing THA prior to LSF were more likely to fill opioid prescriptions compared with patients undergoing THA following LSF at 1 month (aOR 1.63, P < 0.0001 vs. aOR 1.36, P < 0.0001), 3 months (aOR 1.80, P < 0.0001 vs. aOR 1.43, P < 0.0001), 6 months (aOR 2.69, P < 0.0001 vs. aOR 1.35, P < 0.0001), and 12 months (aOR 3.28, P < 0.0001 vs. aOR 1.34, P < 0.0001) following THA (Table 3).
A higher proportion of patients who underwent THA prior to LSF went on to require a revision surgery (9.0%) compared with those undergoing THA following LSF (7.6%, P < 0.0001) (Table 3). Following multivariable regression, risk of revision surgery was higher for patients who underwent THA prior to LSF (aOR 1.91, 95% CI: 1.57–2.30, P < 0.0001) compared with those who underwent THA following LSF (aOR 1.60, 95% CI: 1.34–1.89, P < 0.0001).
In this investigation, patients undergoing THA with a remote history of LSF at least 2 years prior was found to not have increased rates of dislocation rates and 30-day complications but increased rates of revision. For patients with concurrent lumbar spine pathology, those who underwent THA prior to LSF experienced significantly higher rates of dislocation, revision, and prolonged opioid use compared with those who underwent THA following LSF. Therefore, leaving underlying lumbar spinal pathology unaddressed prior to THA was associated with higher complication and revision risks than if a fusion was performed prior to THA. This data may be helpful for patient counseling for those with dual hip and lumbar spine pathology who are considering surgery for both conditions.
To our knowledge, no previous studies have investigated outcomes of THA preceding LSF in patients with dual hip/spine pathology. The present study is the first to report that receiving THA without first addressing an underlying lumbar spinal pathology with LSF leads to higher risk of dislocation, joint prosthetic infection, surgical site infection, revision, and prolonged opioid use than if the THA follows LSF. Prior investigations have shown the THA patients with underlying lumbar spinal pathology, such as lumbar spinal stenosis, are at higher risk for continued pain and disability, poorer outcomes, and revision.12–14 Mahure et al13 prospectively followed patients who had undergone THA with concurrent lumbar spine pathology from a single-surgeon database from 2010 to 2014. Across the study period, patients in the spine pathology copathology cohort faced a revision/THA conversion rate of 23.91%, almost twice as the 11.57% of those in the THA only cohort (P = 0.045). These rates of revision were considerably higher than that of the patients in our study's cohorts, most likely because the database in our study was over many providers versus the single-surgeon database in the study by Mahure et al. Though like Mahure et al, our results also showed an about two-fold increase in risk of revision for THA patients with unaddressed lumbar spine pathology.
With regards to revision, it is important to focus on dislocation, which is the second most common mechanical cause of revision following primary total hip replacement, at 12.2% of revisions.15 A study by King et al16 using the Medicare portion of PearlDiver compared THA patients undergoing subsequent LSF with those who did not and found increased risk of dislocation in the THA patients who received subsequent LSF. In survival analysis, this increased risk persisted 10 years postoperatively, notably with no sudden increase in dislocation risk in the perioperative period following the LSF. In the present study, THA patients who underwent LSF after THA had greater risk of prosthetic dislocation, joint prosthetic infection, and surgical site complication compared with not only the THA only cohort but most importantly also the cohort that underwent LSF prior to THA. Thus, a concurrent diagnosis of spine pathology not first addressed with LSF may play a role in postoperative hip instability following THA. Our findings of increased dislocation risk with LSF following THA in patients with concurrent spine pathology may be attributable to increased stiffness and functional limitations that accompany not only lumbar degenerative disease but also spinal deformity.17
The biomechanical implications of spine-pelvis-hip motion imply that changes in mobility of the spine, either due to pathology or fusion, can affect balance and alignment leading to increased dislocations.7 Studies have demonstrated that lumbar spinal pathologies such as lumbar degenerative disc disease and facet spondylosis of the lumbar spine lead to spinopelvic stiffness, necessitating compensatory increased flexion and extension of the femoral side of the hip in the sitting-standing movement.7,18–20 Esposito et al20 found that patients with degenerative disc disease had decreased spine flexion and increased femoroacetabular flexion in the standing to sitting motion. This may increase the risk of impingement between the prosthetic femoral neck and the edge of the acetabular bone or cup. Impingement is the most common mechanism of dislocation, with another causative factor being increased anteversion, which can lead to posterior impingement leading to anterior dislocation.21 A study of 33 patients found that spinal correction of adult spinal deformity significantly reduced acetabular anteversion by a mean change of −4.96°.22 It is thus advantageous to correct the abnormal spine-pelvis-hip motion in lumbar spinal pathologies first before THA.
THA complications in patients who had recently received LSF have been previously studied. Buckland et al4 retrospectively studied a population of patients who underwent THA with LSF within 1 year prior in relation to patients who underwent THA without spinal fusion. Specifically for dislocation rates, the authors found 2.77 times and 1.93 times greater odds of dislocation for THA following long fusion and short fusion respectively, relative to THA without spinal fusion. However, in our analysis of patients with specifically a remote history of LSF at least 2 years prior to THA, there was not a statistically higher risk of dislocation.
Buckland et al, Blizzard et al, and Sing et al, which report higher prosthetic-related complications and revisions for THA patients having prior LSF, do not adjust for known dislocation risk factors or only adjust for age and sex.4,5,23 In contrast, our study implements additional controls. Perfetti et al6 sought to control for underlying patient risk by performing a propensity score match on patients in New York State. We adjusted for similar underlying variables and similar to Perfetti et al found higher risk of revision in THA patients with prior LSF. A systematic review with subsequent meta-analysis of the resulting six articles reported increased risk of dislocation (RR 2.03) and revision (RR 3.36) as well as poorer patient-reported outcomes in THA patients with prior LSF.24 These patient-reported outcome measures included pain, found to be higher in THA following lumbar surgery.25
Accordingly, opioid utilization following THA in our study found that a greater proportion of THA patients with remote prior LSF were prescribed opioids from 1 month through 12 months postsurgery as compared with other patients. Barry et al found that even in the immediate days following THA surgery, those having received prior LSF demonstrated higher average early postoperative pain scores and total pain medication usage from postoperative day 0 to postoperative day 4.3 Importantly, increased opioid use following orthopedic surgery has been associated with risk of delirium, pneumonia, overdose, hyperalgesia, and fall-related fractures.26–29 This may partially contribute to the increased risk of revisions seen in our population of THA patients with remote prior LSF.
We found that patients with unaddressed lumbar spinal pathology experienced elevated opioid utilization past 3 months postsurgery over those with corrected spinal pathology. Persisting full body pain is most predictive of opioid utilization to 6 months following THA surgery and not localized joint pain.30 Thus, resolving lower back pain first with spinal fusion would likely lead to decreased requirement of opioid following the THA as we found in our study. Though Parvizi et al in a study of 344 patients found that THA surgery alone was able to resolve or improve lower back pain, in light of our opioid utilization results, correcting lower back pain, neurogenic claudication, and radiculopathy with LSF likely affords a more complete general body pain relief than solely relying on THA.31 The lowest opioid utilization is in the group with no lumbar spine problems at all, as THA patients who only received THA had the lowest opioid utilization following surgery. This corroborates Prather et al,12 who found that patients receiving only THA had greater improvements in pain following surgery compared with THA patients also previously diagnosed with a lumbar spinal disorder.
This study has potential limitations. First, in retrospective database studies, data accuracy is contingent on coding accuracy within the system by administrators and physicians. These codes do not capture the severity or symptomology of complications or lumbar spinal pathologies, patient-reported outcomes, or functional outcomes. Further, CPT coding for surgical procedures cannot capture subtle variability in surgical technique such as THA approach. Next, the PearlDiver database does not report statistical results for groups fewer than 11 patients. Thus, the risks of low incidence complications were not reported, despite being included in the measure of total complications. Additionally, because PearlDiver is a private-payer insurance database, the external validity of the findings of this study may not hold true for non-Humana patients or patients on public insurance. Lastly, in the opioid utilization analysis, by 1 year out from surgery, a patient may have received another procedure, inflating the risk of opioid utilization following the index surgery.
Despite these limitations, the findings of this study are potentially valuable given that these data help to inform patients who have lumbar spine and hip pathology that require surgery. This investigation revealed that leaving lumbar spinal pathology unaddressed before THA was more detrimental to THA outcomes than if an LSF preceded the THA.
Spine and joint replacement surgeons should coordinate patient care in individuals with dual hip and spine pathology. To best optimize surgical outcomes and pain control, patients seeking both LSF and THA may consider addressing their spinal pathology first as complications following THA may be lower when LSF is performed prior to THA as compared with when THA is performed prior to LSF.
- Compared with patients only receiving THA, THA patients with remote prior LSF experienced increased risk of revision as well as increased likelihood of prolonged postoperative opioid use.
- For patients with concomitant spine and hip pathology, those who underwent THA prior to LSF had increased dislocation risk, higher rate of prosthetic joint infection, surgical site complications, revision, and postoperative opioid use compared with those who had THA following LSF.
- To best optimize surgical outcomes and pain control, patients seeking both LSF and THA may consider addressing their spinal pathology first.
1. Maradit Kremers H, Larson DR, Crowson CS, et al. Prevalence of total hip and knee replacement in the United States. J Bone Jt Surgery-American Vol
2. Rivière C, Lazennec J-Y, Van Der Straeten C, et al. The influence of spine-hip relations on total hip replacement: a systematic review. Orthop Traumatol Surg Res
3. Barry JJ, Sing DC, Vail TP, et al. Early outcomes of primary total hip arthroplasty
after prior lumbar spinal fusion
. J Arthroplasty
4. Buckland AJ, Puvanesarajah V, Vigdorchik J, et al. Dislocation
of a primary total hip arthroplasty
is more common in patients with a lumbar spinal fusion
. Bone Joint J
5. Blizzard DJ, Sheets CZ, Seyler TM, et al. The impact of lumbar spine disease and deformity on total hip arthroplasty
6. Perfetti DC, Schwarzkopf R, Buckland AJ, et al. Prosthetic dislocation
and revision after primary total hip arthroplasty
in lumbar fusion patients: a propensity score matched-pair analysis. J Arthroplasty
7. Ike H, Dorr LD, Trasolini N, et al. Spine-pelvis-hip relationship in the functioning of a total hip replacement. J Bone Jt Surg
8. Pugely AJ, Bedard NA, Kalakoti P, et al. Opioid use
following cervical spine surgery: trends and factors associated with long-term use. Spine J
9. Gause TM, Nunnery JJ, Chhabra AB, et al. Perioperative narcotic use and carpal tunnel release: trends, risk factors, and complications
. HAND (N Y)
2018; 155894471879227[Epub ahead of print].
10. Johnson SP, Chung KC, Zhong L, et al. Risk of prolonged opioid use
among opioid-naïve patients following common hand surgery procedures. J Hand Surg Am
11. Steiner SRH, Cancienne JM, Werner BC. Narcotics and knee arthroscopy: trends in use and factors associated with prolonged use and postoperative complications
12. Prather H, Van Dillen LR, Kymes SM, et al. Impact of coexistent lumbar spine disorders on clinical outcomes and physician charges associated with total hip arthroplasty
. Spine J
13. Mahure SA, Ryan MK, Buckland A, et al. Concomitant lumbar spine pathology in patients undergoing hip arthroscopy: a matched cohort analysis. Orthop J Sport Med
2017; 5: (7_suppl6): 2325967117S0042.
14. Bohl WR, Steffee AD. Lumbar spinal stenosis. A cause of continued pain and disability in patients after total hip arthroplasty
. Spine (Phila Pa 1976). 4(2):168-173. Available at: http://www.ncbi.nlm.nih.gov/pubmed/264032
. Accessed February 7, 2019.
15. Bozic KJ, Lau E, Ong K, et al. Risk factors for early revision after primary total hip arthroplasty
in medicare patients. Clin Orthop Relat Res
16. King CA, Landy DC, Martell JM, et al. Time to dislocation
analysis of lumbar spine fusion following total hip arthroplasty
: breaking up a happy home. J Arthroplasty
17. Daniels AH, Smith JS, Hiratzka J, et al. Functional limitations due to lumbar stiffness in adults with and without spinal deformity. Spine (Phila Pa 1976)
18. Gorin M, Roger B, Lazennec J-Y, et al. Hip-spine relationship: a radio-anatomical study for optimization in acetabular cup positioning. Surg Radiol Anat
19. Lazennec J-Y, Boyer P, Gorin M, et al. Acetabular anteversion with CT in supine, simulated standing, and sitting positions in a tha patient population. Clin Orthop Relat Res
20. Esposito CI, Miller TT, Kim HJ, et al. Does degenerative lumbar spine disease influence femoroacetabular flexion in patients undergoing total hip arthroplasty
? Clin Orthop Relat Res
21. Zahar A, Rastogi A, Kendoff D. Dislocation
after total hip arthroplasty
. Curr Rev Musculoskelet Med
22. Buckland AJ, Vigdorchik J, Schwab FJ, et al. Acetabular anteversion changes due to spinal deformity correction: bridging the gap between hip and spine surgeons. J Bone Joint Surg Am
23. Sing DC, Barry JJ, Aguilar TU, et al. Prior lumbar spinal arthrodesis increases risk of prosthetic-related complication in total hip arthroplasty
. J Arthroplasty
24. An VVG, Phan K, Sivakumar BS, et al. Prior lumbar spinal fusion
is associated with an increased risk of dislocation
and revision in total hip arthroplasty
: a meta-analysis. J Arthroplasty
25. Eneqvist T, Nemes S, Brisby H, et al. Lumbar surgery prior to total hip arthroplasty
is associated with worse patient-reported outcomes. Bone Joint J
26. Menendez ME, Ring D, Bateman BT. Preoperative opioid misuse is associated with increased morbidity and mortality after elective orthopaedic surgery. Clin Orthop Relat Res
27. Kidner CL, Mayer TG, Gatchel RJ. Higher opioid doses predict poorer functional outcome in patients with chronic disabling occupational musculoskeletal disorders. J Bone Joint Surg Am
28. Chu L, Clark D, Angst M. Opioid tolerance and hyperalgesia in chronic pain patients after one month of oral morphine therapy: a preliminary prospective study. J Pain
29. Saunders KW, Dunn KM, Merrill JO, et al. Relationship of opioid use
and dosage levels to fractures in older chronic pain patients. J Gen Intern Med
30. Goesling J, Moser SE, Zaidi B, et al. Trends and predictors of opioid use
after total knee and total hip arthroplasty
31. Parvizi J, Pour AE, Hillibrand A, et al. Back pain and total hip arthroplasty
: a prospective natural history study. Clin Orthop Relat Res