- Email: [email protected]
The impact of preoperative depression on quality of life outcomes after lumbar surgery
The Spine Journal 15 (2015) 58–64
Clinical Study
The impact of preoperative depression on quality of life outcomes after lumbar surgery Jacob A. Miller, BSa,b, Adeeb Derakhshan, BSa,b, Daniel Lubelski, BAa,b, Matthew D. Alvin, MBA, MAa,c, Matthew J. McGirt, MDd, Edward C. Benzel, MDa,b,e, Thomas E. Mroz, MDa,b,e,* a
Cleveland Clinic Center for Spine Health, Cleveland Clinic, 9500 Euclid Ave., S40, Cleveland, OH 44195, USA b Cleveland Clinic Lerner College of Medicine, 9500 Euclid Ave., NA41, Cleveland, OH 44195, USA c Case Western Reserve University School of Medicine, 10900 Euclid Ave., Cleveland, OH 44106, USA d Carolina Neurosurgery and Spine Associates, 225 Baldwin Ave., Charlotte, NC 28204, USA e Department of Neurological Surgery, Cleveland Clinic, 9500 Euclid Ave., S40, Cleveland, OH 44195, USA Received 25 November 2013; revised 13 April 2014; accepted 30 June 2014
Abstract
BACKGROUND CONTEXT: Some, smaller studies have investigated the effect of preoperative depression on postoperative improvement in quality of life (QOL). However, they have not used the Patient Health Questionnaire 9 (PHQ-9) in self-reported depression. PURPOSE: To assess the effect of preoperative depression as measured by the PHQ-9 on postoperative improvement in QOL. STUDY DESIGN: A retrospective review at a single tertiary-care referral center. PATIENT SAMPLE: Patients who underwent lumbar decompression or fusion between 2008 and 2012. OUTCOMES MEASURES: A self-reported EuroQol five-dimensions (EQ-5D) quality-adjusted life-years Index. METHODS: Quality of life data were collected using the institutional prospectively collected database of patient-reported health status measures. The EQ-5D questionnare, PDQ, and PHQ-9 were used. Linear and logistic regression analyses were performed to assess the impact of preoperative depression on QOL improvement. RESULTS: Elevated preoperative pain (PDQ, b50.0017, p5.0009) and worsened depression (PHQ-9, b50.0044, p5.0359) were significantly associated with diminished postoperative improvement in QOL, as measured by the EQ-5D. Furthermore, greater depression (PHQ-9, odds ratio [OR] 0.93, p!.0001) and pain (PDQ, OR 0.99, p5.02) were associated with significantly diminished postoperative improvement exceeding the minimum clinically important difference. CONCLUSIONS: Increased preoperative pain and depression were shown to be associated with significantly reduced improvement in postoperative QOL, as measured by the EQ-5D. Ó 2015 Elsevier Inc. All rights reserved.
Keywords:
Lumbar surgery; EQ-5D; PHQ-9; PDQ; Depression; Psychosocial outcomes
FDA device/drug status: Not applicable. Author disclosures: JAM: Nothing to disclose. AD: Nothing to disclose. DL: Nothing to disclose. MDA: Nothing to disclose. MJM: Grants: Depuy and Stryker Spine (E, paid directly to institution). ECB: Royalties: Elsevier Pub (B), Thieme Pub (B); Stock Ownership: Axiomed (NA), Depuy (NA), Orthomems (NA), Turning Point (NA); Consulting: Axiomed; Speaking and or Teaching Arrangements: Multiple (Varied); Trips/Travel: Multiple (Varied); Grants: Orthopaedic Research and Education Foundation (F, Paid directly to institution), Rawlings (F, Paid directly to institution). TEM: Stock Ownership: PearlDiver, Inc. (No money received); http://dx.doi.org/10.1016/j.spinee.2014.06.020 1529-9430/Ó 2015 Elsevier Inc. All rights reserved.
Consulting: Globus Medical (B); Speaking and/or Teaching Arrangements: AO Spine (B). The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. IRB approval (IRB #11-957) was obtained before initiation of the study. No funding sources or conflict of interests were present. * Corresponding author. Cleveland Clinic Center for Spine Health, Neurological Institute, The Cleveland Clinic, 9500 Euclid Ave., S80, Cleveland, OH 44195, USA. Tel.: (216) 445-9232; fax: (216) 363-2040. E-mail address: [email protected] (T.E. Mroz)
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
59
Introduction In appropriate patients, lumbar decompression and fusion significantly improve the quality of life (QOL) [1–5]. However, such surgical procedures are associated with significant cost and harbor the potential for significant complications [4,6–11]. In an effort to reduce costs and improve the QOL outcomes after spine surgery, researchers have investigated the preoperative risk factors to better identify optimal surgical candidates. These studies have preliminarily identified the risk factors related to comorbidities, demographic, socioeconomic, and psychological factors. Diabetes and smoking, for example, have been shown to increase the incidence of complications and length of stay after lumbar operations for lumbar canal stenosis, prolapsed discs, and thoracolumbar scoliosis [12]. A systematic review of the literature identified a lower level of education, higher preoperative pain level, low work satisfaction, and longer duration of sick leave as risk factors for unfavorable outcomes, as they relate to pain, disability, and return to work [13]. Prior studies have identified a strong correlation between psychological factors, such as depression, and poor clinical outcomes after lumbar spine surgery, as measured by the patient-reported metrics such as the visual analog scale (VAS) and Oswestry Disability Index [4,6–11]. A systematic review of the literature by Celestin et al. [9] noted that preoperative depression, anxiety, and poor coping abilities were associated with reduced surgical benefit after lumbar surgical procedures. Other socioeconomic factors, such as workers’ compensation litigation, have also been implicated in worsened postoperative QOL outcomes [14,15]. These findings support the use of preoperative psychosocial questionnaires to better predict the patients most likely to achieve optimal postoperative QOL outcomes. Although other studies have investigated the effect of preoperative depression on QOL outcomes, none have used the Patient Health Questionnaire 9 (PHQ-9) that has demonstrated the highest sensitivity and specificity in assessing self-reported depression among psychosocial questionnaires [16–18]. Furthermore, no studies have used the Pain Disability Questionnaire (PDQ) in conjunction with depression questionnaires to evaluate the effects of pain or preoperative QOL on postoperative outcomes [19]. In the present study, we sought to assess the impact of preoperative depression, and other baseline patient characteristics, on QOL outcomes after lumbar surgery. We hypothesized that lumbar spine surgery significantly improves the QOL outcomes in all patients, but with a statistically significant lower improvement in those with greater preoperative depression.
Context The negative influence of behavioral health conditions on outcomes after surgical intervention has been surmised on a number of levels. The authors sought to add to this literature by reviewing patient experiences following elective lumbar spine surgery using a variety of quality of life and mental health questionnaires, including the Patient Health Questionnaire-9 (PHQ-9). Contribution The authors evaluated results among a series of 919 patients. Elevated levels of preoperative pain and depression (as measured on the PHQ-9) were found to adversely influence postoperative improvement in quality of life as reflected on the EQ-5D. Implications The authors’ analysis adds to a growing body of literature that improves understanding regarding the influence of behavioral health conditions on outcomes after spine surgery. This study is remarkable due to the amount of data collected on participating patients and the number of individuals included. It remains a retrospective review, however, and one that was performed at a single tertiary center. As a result, the findings may not be generalizable to all individuals encountered in spine surgical practices throughout the nation. —The Editors
between 2008 and 2012, at a single tertiary-care center for nontraumatic and nonneoplastic indications. Quality of life data were collected using the institutional prospectively collected database of patient-reported health status measures, called the Knowledge Program that includes validated questionnaires such as the EuroQol five-dimensions (EQ-5D), PDQ, and PHQ-9, administered at each pre- and postoperative outpatient visit. Patients included in this study were followed for a minimum of 6 months within the study period, with one of the following diagnoses: lumbosacral spondylosis, degenerative disc disease, lumbar spinal stenosis, lumbar disc displacement, lumbar kyphosis, or lumbar scoliosis. Patients were excluded if they were younger than 18 years at the time of surgery, had follow-up of less than 6 months, or had incomplete pre- or postoperative QOL data. QOL outcome measures
Methods Patient selection A retrospective review was performed of all patients who underwent elective lumbar decompression and fusion
Health-status measures questionnaires (PDQ, PHQ-9, and EQ-5D) were collected in a prospective manner, preoperatively, and at each subsequent follow-up visit. The previously validated [19,20] PDQ records functional and psychosocial components of pain, with a total score between 0 and 150, with increasing pain reflected by an
60
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
increasing score. The minimum clinically important difference (MCID) in PDQ has been reported to be 20, although this has not been studied in a lumbar surgery population; rather, it has been studied in a chronic pain population [21]. The PHQ-9 was used for the assessment of depression. This questionnaire assesses for depression based on nine Diagnostic and Statistical Manual of Mental Disorders V [22] criteria for major depressive disorder. Its nine questions are scored from 0 (not at all) to 3 (nearly every day), making the total score range from 0 to 27. The total score is commonly categorized into one of five groups: no symptoms (score50–4), minimal symptoms (score55–9), minor depression (score510–14), moderately severe major depression (score515–19), and severe major depression (score520–27). The diagnostic validity of the PHQ-9 has been established in large multicenter analyses [5,17] and in a chronic spinal pain population [16]. The MCID for the PHQ-9 has been reported to be 5, although this has not been studied in a lumbar surgery population; rather, it has been studied in depressed patients being treated in psychiatric clinics [23]. The EQ-5D contains five dimensions of health state: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension is scored from 0 to 3. For example, for the anxiety/depression dimension, patients check a box indicating that they are ‘‘not anxious or depressed,’’ ‘‘moderately anxious or depressed,’’ or ‘‘extremely anxious or depressed,’’ which would respectively correlate with increasing scores [24–26]. An EQ5D index score is then calculated from these five dimension subscores. This index score represents the patient’s qualityadjusted life-years (QALYs), ranging from a QALY of 0 (death) to 1 (perfect health), and is included in the analysis as a reflection of the patient’s overall QOL. As such, the EQ-5D index represents a calculated QALY. The MCID for the EQ-5D after surgery has been reported to be 0.1 [4]. Once identified, the patient sample was assessed for the impact of depression (via PHQ-9) on postoperative QOL outcomes.
regression analysis was performed using postoperative EQ-5D QALY change exceeding the MCID of 0.14 as the dependent variable. Again, variables that were associated (p!.2) via simple logistic regression were included in the multiple logistic regression. Regression coefficients (b) were reported for linear regression, and odds ratios (ORs) with 95% confidence intervals were reported for logistic regression. Data were analyzed using the JMP pro 10 statistical software (SAS Institute, Inc., Carey, NC, USA, 2012). For comparisons with multiple t tests, a Bonferroni correction was made, and thus, p values!.01 were considered statistically significant. Otherwise, p values!.05 were considered statistically significant.
Results Demographics Two thousand nine hundred fifty-seven patients were eligible for inclusion. Of these, 1,975 (67%) had a follow-up of less than 6 months and 63 (2%) did not have pre- or postoperative QOL data, and were thus excluded. Hence, 919 patients were included in the study. The cohort had an average follow-up of 421 days. Four hundred sixty-eight (50.9%) patients were men and 833 (90.6%) were Caucasian (Table 1). The average body mass index was 29.1, Table 1 Baseline presentation of 919 patients who underwent lumbar surgery Baseline
n5919
Age (y) Male gender BMI Race
59.6613.7 468 (50.9%) 29.165.8 Asian: 2 (0.2%) Black: 58 (6.3%) Caucasian: 833 (90.6%) Mixed: 4 (0.5%) Other/unknown: 22 (2.4%) Renal failure: 20 (2.2%) Diabetes: 153 (16.7%) Coronary artery disease: 78 (8.5%) Hypertension: 356 (38.8%) Stroke history: 17 (1.9%) $53,399 272 (29.6%) Lumbosacral spondylosis: 118 (12.8%) Degenerative disc disease: 84 (9.1%) Spinal stenosis: 91 (9.9%) Disc displacement: 515 (56.2%) Lumbar kyphosis: 28 (3.0%) Lumbar scoliosis: 83 (9.0%) Laminotomy: 239 (26.0%) Laminoplasty: 15 (1.6%) Laminectomy: 174 (18.9%) Anterior fusion: 12 (1.3%) Posterolateral fusion: 267 (29.1%) Posterior fusion: 212 (23.1%) 421.26216.7 179 (19.5%)
Comorbidities
Statistical analysis The primary aim of this study was to identify if there was an association between preoperative depression (as assessed by PHQ-9) and postoperative improvement in EQ5D QALY scores. Continuous data were expressed as mean6standard deviation and compared via the Student t test; paired data (eg, preoperative and postoperative QALYs) were compared with a paired t test. Categorical data were compared with the Fisher Exact test. To avoid overfitting the model, only variables that demonstrated association (p!.2) with improvements in EQ-5D QALY (via simple linear regression) were analyzed via a multiple linear regression model to assess for independent predictors of improved QOL outcomes. Preoperative EQ-5D QALY was included in the model to adjust for the ceiling effect and normalize all improvements to baseline scores. A logistic
Income (median) Antidepressants Diagnosis
Surgery type
Follow-up (d) Depressed (PHQ-9O14)
BMI, body mass index; PHQ-9, patient health questionnaire 9.
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
and the median annual income was $53,399. A significant portion of the population had comorbidities at the time of surgery: 20 (2.2%) had chronic renal failure, 153 (16.7%) had diabetes, 78 (8.5%) had coronary artery disease, 356 (38.8%) had hypertension, and 17 (1.9%) had a history of cerebral infarction. There were 272 (29.6%) patients taking antidepressants and 179 (19.5%) reporting moderate or severe depression (PHQ-9 score greater than 14). Four hundred ninety-one (53.5%) patients underwent lumbar fusion procedures, whereas the remaining 428 (46.5%) underwent lumbar decompression without fusion. Anterior interbody fusions accounted for 1.3% of the study population, whereas posterolateral and posterior interbody fusions accounted for 29.1% and 23.1%, respectively. The indications for these procedures varied, although more than half (56.2%) were for diagnoses of disc displacement. Postoperative health-related QOL outcomes At the last follow-up, statistically significant improvements in QOL outcomes in all questionnaires were observed in the cohort (Table 2). The following pre- and postoperative averages were observed: PDQ functional component (55.2 preoperatively vs. 34.4 postoperatively), PDQ psychosocial component (31.0 preoperatively vs. 19.1 postoperatively), PDQ total (86.3 preoperatively vs. 53.4 postoperatively), PHQ-9 (9.8 preoperatively vs. 5.1 postoperatively), and EQ-5D QALY (0.49 preoperatively Table 2 Pre- and postoperative quality of life outcomes Quality of life measure PDQ FC Preoperative Postopoperative Change PDQ PC Preoperative Postoperative Change PDQ total Preoperative Postoperative Change PHQ-9 Preoperative Postoperative Change EQ-5D index Preoperative Postoperative Change EQ-5D QALYOMCID (0.1) PHQ-9OMCID (5) PDQOMCID (20)
Lumbar surgeries (n5919)
p
55.2618.0 34.4625.7 20.3625.1
!.0001
31.0613.1 19.1616.6 11.7614.9
!.0001
86.3629.4 53.4641.2 32.2638.5
!.0001
9.866.8 5.165.6 3.566.3
!.0001
0.4960.22 0.6560.22 0.1560.25 461 (50.2%) 347 (37.8%) 532 (57.9%)
!.0001
— — —
EQ-5D, EuroQol five-dimensions; PDQ, pain disability questionnaire; FC, functional component; PC, psychosocial component; PHQ-9, patient health questionnaire 9; QALY, quality-adjusted life-years; MCID, minimum clinically important difference. Note: Paired Student t test was used to check for significance (p!.01).
61
vs. 0.65 postoperatively) (p!.0001 for all questionnaires). Four hundred sixty-one (50.2%) patients experienced improvement exceeding the MCID of 0.1 in EQ-5D QALY. Additionally, 347 (37.9%) and 532 (57.9%) patients experienced improvement exceeding the MCID of 5 in PHQ-9 and of 20 in PDQ, respectively. Regression analysis Simple and multiple linear regression analyses were performed to assess the effect of preoperative depression, as measured by the PHQ-9 on postoperative QOL improvement (Table 3). For both of these analyses, EQ-5D QALY improvement (postoperative minus preoperative QALYs) was used as the dependent outcome variable. Simple linear regression revealed six variables that were moderately associated (p!.2) with improvement: age (b50.0006, p5.1866), procedures with fusion (b50.0356, p5.0032), procedure type (b50.6058, p5.0050), preoperative EQ5D QALY (b50.6389, p!.0001), preoperative PHQ-9 score (b50.0083, p!.0001), and preoperative PDQ score (b50.0021, p!.0001). Both ‘‘procedure type’’ and ‘‘procedures with fusion’’ were included as separate variables in the model to test for the independent effect of fusion. These six variables were included in a multiple linear regression model. After modeling, age (b50.0008, p5.2315), procedure type (b50.0612, p5.79), and lumbar fusion (b50.0006, p5.6344) were no longer significant predictors of EQ-5D QALY improvement. The remaining three variables (preoperative EQ-5D QALY, PHQ-9, and PDQ scores) Table 3 Linear regression analysis of independent preoperative predictors of improvement in health state (change in EQ-5D score) after lumbar surgery Simple regression
Multiple regression
Variables
Coefficient
p
Coefficient
Age (y) Race* Female gender Income BMI Lumbar fusion Procedure typey Diagnostic indicationz Renal failure Diabetes Coronary artery disease Hypertension Stroke history Antidepressant use EQ-5D QALY baseline PHQ-9 baseline PDQ baseline
0.0006 0.0018 0.0048 !0.0001 0.0001 0.0356 0.6058 0.0017 0.0094 0.0002 0.0096 0.0028 0.0206 0.0025 0.6389 0.0083 0.0021
.1866 .8983 .5652 .2825 .7642 .0032 .0050 .5501 .7380 .9865 .5185 .7423 .5014 .7836 !.0001 !.0001 !.0001
p
0.0008
.2315
0.0006 0.0612
.6344 .7906
0.8915 0.0044 0.0017
!.001 .0359 .0009
EQ-5D, EuroQol five-dimensions; QALY, quality-adjusted life-years; PHQ-9, patient health questionnaire 9; PDQ, pain disability questionnaire; BMI, body mass index. * Coefficient reported for Caucasian. y Coefficient reported for laminectomy. z Coefficient reported for disc displacement.
62
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
were all found to be significant independent predictors of EQ-5D QALY improvement. The regression coefficients were all negative, indicating that increasing preoperative pain (PDQ, b50.0017, p5.0009) and worsening PHQ-9 depression index (PHQ-9, b50.0044, p!.0359) reduced postoperative improvement. The normalization variable, EQ-5D QALY, was also a significant predictor in the model (b50.8915, p!.001). Simple and multiple logistic regression analyses were also performed to assess the effect of preoperative PHQ-9 depression index on postoperative QOL improvement exceeding the MCID (Table 4). For both of these analyses, EQ-5D QALY improvement exceeding the MCID (0.1) was used as the dependent outcome variable. Simple logistic regression identified seven variables that were moderately associated (p!.2) with EQ-5D QALY exceeding the MCID: performance of lumbar fusion (OR 1.62; 95% confidence intervals: [1.10, 2.38], p5.01), procedure type (OR 0.20; [0.15, 0.80], p5.002), diagnostic indication (OR 0.96; [0.92, 1.01], p5.10), history of stroke (OR 2.44; [0.90, 7.73], p5.08), preoperative EQ-5D QALY (OR 0.005; [0.002, 0.01], p!.0001), preoperative PHQ-9 (OR 1.06; [1.04, 1.09], p!.0001), and preoperative PDQ (OR 1.02; [1.01, 1.03], p!.0001). These seven variables were included in a multiple logistic regression. After modeling, only preoperative EQ-5D QALY (OR 0.001; [0.0003, 0.002], p!.0001), PHQ-9 (OR 0.93; [0.90, 0.96], p!.0001), and PDQ (OR 0.99; [0.97, 0.99], p5.02) remained independent significant predictors of QALY change exceeding the MCID. The ORs for these three variables were all less than 1, indicating that worsening PHQ-9 depression index
(PHQ-9, OR 0.93, p!.0001) and pain (PDQ, OR 0.99, p5.02) reduced postoperative improvement exceeding the MCID. The normalization variable, EQ-5D QALY, was also a significant predictor in the model (OR 0.001, p!.0001).
Discussion In the present study, we sought to investigate the effect of preoperative depression as measured by the PHQ-9 on postoperative QOL through the use of three patient reported questionnaires (PDQ, PHQ-9, and EQ-5D). Multiple linear regression showed that each of these health status measures was, individually, a significant predictor of improvement in EQ-5D QALY, with greater PHQ-9 depression index and pain being predictive of worse postoperative QOL outcomes. Although the present study is the first to use the PDQ and PHQ-9 in postoperative QOL assessment after lumbar surgery, several studies have used other tools in assessing the association between depression and outcome. Arpino et al. [27] used the Zung depression scale in their prospective investigation of 73 patients who underwent microdiscectomy for lumbar disc herniation. The authors concluded that patients with lower preoperative Zung depression scores (less depression) had a better outcome in pain (as measured by VAS) at 3 and 12 months after surgery. In a prospective clinical study, Sinikallio et al. [28] used the Beck Depression Inventory (BDI) to assess the 102 patients who underwent decompression for lumbar spinal stenosis. They reported that higher preoperative BDI scores were independently
Table 4 Logistic regression analysis of independent preoperative predictors of MCID Simple regression
Multiple regression
Variables
OR (95% CI)
p
Age (y) Race* Female gender Income BMI Lumbar fusion Procedure typey Diagnostic indicationz Renal failure Diabetes Coronary artery disease Hypertension Stroke history Antidepressant use EQ-5D QALY baseline PHQ-9 baseline PDQ baseline
1.57 1.33 1.07 1.01 0.98 1.62 0.20 0.96 0.66 0.90 0.82 0.98 2.44 1.16 0.005 1.06 1.02
.20 .20 .62 .98 .50 .01 .002 .10 .37 .54 .41 .86 .08 .30 !.0001 !.0001 !.0001
(0.78, 3.15) (0.86, 2.09) (0.82, 1.38) (0.43, 0.99) (0.93, 1.04) (1.10, 2.38) (0.15, 0.80) (0.92, 1.01) (0.26, 1.61) (0.63, 1.27) (0.51, 1.31) (0.75, 1.28) (0.90, 7.73) (0.87, 1.55) (0.002, 0.01) (1.04, 1.09) (1.01, 1.03)
OR (95% CI)
p
1.08 (0.52, 2.30) 0.80 (0.56, 1.25) 0.97 (0.90, 1.04)
.83 .65 .36
1.00 (0.99, 1.01)
1.00
0.001 (0.0003, 0.002) 0.93 (0.90, 0.96) 0.99 (0.97, 0.99)
!.0001 !.0001 .02
MCID, minimum clinically important difference; OR, odds ratio; CI, confidence interval; BMI, body mass index; EQ-5D, EuroQol five-dimensions; QALY, quality-adjusted life-years; PHQ-9, patient health questionnaire 9; PDQ, pain disability questionnaire. * Coefficient reported for Caucasian. y Coefficient reported for laminectomy. z Coefficient reported for disc displacement.
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
predictive of poorer self-reported functional ability, symptom severity, and poorer walking capacity at 1-year follow-up. Chaichana et al. [8] used the Zung self-rating depression scale in their prospective study of 67 patients who underwent lumbar discectomy. They determined that a patient with higher preoperative depression was significantly less likely to achieve an MCID in disability or QOL after surgery, thereby reaffirming the predictive value of preoperative depression in postoperative outcome. Adogwa et al. [10] used the Zung depression scale to retrospectively review the outcomes of 150 patients who underwent revision lumbar surgery for symptomatic pseudarthrosis, adjacent segment disease, and same-level recurrent stenosis. They concluded that higher (worse) preoperative Zung depression scores were independently associated with a decreased 2year improvement in QOL outcome in these patients. In a separate study, Adogwa et al. [11] retrospectively investigated 53 patients who underwent revision surgery for same-level lumbar stenosis. Using the Zung self-rating depression scale, the authors evaluated the QOL outcomes (via the EQ-5D, Oswestry Disability Index, and the pain VAS) 2 years after surgery and concluded that increasing (worsening) preoperative Zung depression score was statistically and clinically significantly associated with diminished improvement in QOL outcome postoperatively. Our results corroborate the results of these studies and expand the patient population to include individuals with all types of lumbar decompression or fusion. The studies referenced in the previous paragraph did not use the PHQ-9 measure to assess the impact of depression on postoperative QOL outcomes. The purpose of this study was to, therefore, investigate the effect of preoperative depression using the PHQ-9 on QOL outcomes after decompression and fusion lumbar surgeries in a large cohort. Our initial hypothesis that greater preoperative PHQ-9 depression index would lead to decreased postoperative improvement was confirmed through analysis via multiple linear and logistic regressions. This suggests that the PHQ-9 is at least on par with the Zung self-rating depression scale and the BDI. Following statistical analysis, several significant results were noted. There were significant differences between the preoperative and postoperative PDQ (both the functional and psychosocial components), PHQ-9, and EQ-5D index (p!.0001 for all). This suggests, as expected, that surgery improved the patients’ QOL in general. In the multiple linear regression analysis, the baseline EQ-5D QALY score (p!.001), baseline PHQ-9 score (p5.0359), and baseline PDQ score (p5.0009) were all independently and statistically significant predictive of an improvement in health state (as measured by the change in EQ-5D QALY) after lumbar surgery. Additionally, these variables were also significant predictors of change in EQ-5D QALY exceeding the MCID, demonstrating that these results have clinical and statistical significance. It is important to include the EQ-5D QALY baseline score in the model, as it corrects for all levels of preoperative QOL, and therefore, adjusts for the ceiling
63
effect. Furthermore, inclusion of the variable generalizes the model to include patients with any preoperative EQ5D QALY, whether low or high. Notably, potential confounding factors such as diagnostic indication for surgery were not independently associated with improvement. This study includes the largest cohort of lumbar spine surgery patients to date for the assessment of variables associated with changes in QOL. We assessed for depression using the PHQ-9, based on the Diagnostic and Statistical Manual of Mental Disorders V criteria. This study contributes to the body of evidence that demonstrates the negative impact of preoperative depression on postoperative QOL outcomes. Furthermore, the model is generalizable to surgeries with and without fusion. Previous smaller studies support our findings, although these studies used QOL surveys with reduced sensitivity and specificity in detecting depression [6– 9,11,15,16,27,28]. Although 919 patients were included after exclusion, a significant number of patients (1,975) were excluded because of follow-up of less than 6 months. This loss to follow-up is a potential confounding factor in the data: individuals with a follow-up of less than 6 months may have superior postoperative outcomes independent of depression state, and thus, the observed effects in the present study may pertain only to those with long follow-up times, given the chronic nature of their disease. Furthermore, it is more likely that individuals with long follow-up times are depressed, given the length of time they experience symptoms. The results of this study have significant clinical implications. This study suggests that depression is a modifiable risk factor for poor outcomes. Surgeons should be cognizant of the fact that patients that are preoperatively depressed are more likely to experience poorer outcomes after lumbar surgery. As such, greater attention should be paid to accurately evaluate all patients preoperatively with a validated psychosocial questionnaire. A trained multidisciplinary team may then address and treat depression before progressing to surgical intervention in an effort to maximize the benefit of surgery. Finally, depression should be addressed in the postoperative follow-up in an effort to prevent relapse into depression, as it may diminish the postoperative benefits of surgery. The results of this study should be interpreted with consideration of the study’s limitations. Because of the perceived stigma of depression, patients may not honestly and objectively report their depression. Moreover, the patient’s emotional status may vary between visits and, as such, answers given on the questionnaire may not be an accurate portrayal of how the patient typically feels. Additionally, this was a retrospective study with a relatively short follow-up period of 1 year. Prospectively designed studies with longer follow-ups are needed to further validate the findings. Finally, as this study was conducted at a single tertiary-care institution, the external validity may be limited to institutions with similar patient populations and demographics. Despite these limitations, the methods and assumptions in the present study adhere similarly to
64
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
what has been published previously in the literature, thereby allowing us to justify our conclusions.
Conclusion Worse preoperative pain and depression are significantly associated with diminished improvement in QOL after lumbar surgery. These findings support the use of preoperative psychosocial questionnaires that may serve as a tool for selecting patients more likely to have superior postoperative outcomes and those who are unlikely to derive benefit from the surgery. Patients should be made aware of these associations when deciding on surgical treatment. References [1] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Aaronson O, Cheng JS, et al. Cost per quality-adjusted life year gained of revision neural decompression and instrumented fusion for same-level recurrent lumbar stenosis: defining the value of surgical intervention. J Neurosurg Spine 2012;16:135–40. [2] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Devin CJ, Cheng JS, et al. Cost per quality-adjusted life year gained of laminectomy and extension of instrumented fusion for adjacent-segment disease: defining the value of surgical intervention. J Neurosurg Spine 2012;16:141–6. [3] Adogwa O, Parker SL, Davis BJ, Aaronson O, Devin C, Cheng JS, et al. Cost-effectiveness of transforaminal lumbar interbody fusion for Grade I degenerative spondylolisthesis. J Neurosurg Spine 2011;15:138–43. [4] Parker SL, Adogwa O, Mendenhall SK, Shau DN, Anderson WN, Cheng JS, et al. Determination of minimum clinically important difference (MCID) in pain, disability, and quality of life after revision fusion for symptomatic pseudoarthrosis. Spine J 2012;12:1122–8. [5] Tharin S, Mayer E, Krishnaney A. Lumbar microdiscectomy and lumbar decompression improve functional outcomes and depression scores. Evid Based Spine Care J 2012;3:65–6. [6] Trief PM, Grant W, Fredrickson B. A prospective study of psychological predictors of lumbar surgery outcome. Spine 2000;25:2616–21. [7] LaCaille RA, DeBerard MS, Masters KS, Colledge AL, Bacon W. Presurgical biopsychosocial factors predict multidimensional patient: outcomes of interbody cage lumbar fusion. Spine J 2005;5:71–8. [8] Chaichana KL, Mukherjee D, Adogwa O, Cheng JS, McGirt MJ. Correlation of preoperative depression and somatic perception scales with postoperative disability and quality of life after lumbar discectomy. J Neurosurg Spine 2011;14:261–7. [9] Celestin J, Edwards RR, Jamison RN. Pretreatment psychosocial variables as predictors of outcomes following lumbar surgery and spinal cord stimulation: a systematic review and literature synthesis. Pain Med 2009;10:639–53. [10] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Aaronson OS, Cheng JS, et al. Preoperative Zung depression scale predicts outcome after revision lumbar surgery for adjacent segment disease, recurrent stenosis, and pseudarthrosis. Spine J 2012;12:179–85.
[11] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Bydon A, Cheng JS, et al. Preoperative Zung depression scale predicts patient satisfaction independent of the extent of improvement after revision lumbar surgery. Spine J 2013;13:501–6. [12] Appaduray SP, Lo P. Effects of diabetes and smoking on lumbar spinal surgery outcomes. J Clin Neurosci 2013;20:1713–7. [13] den Boer JJ, Oostendorp RAB, Beems T, Munneke M, Oerlemans M, Evers AWM. A systematic review of bio-psychosocial risk factors for an unfavourable outcome after lumbar disc surgery. Eur Spine J 2006;15:527–36. [14] Carreon LY, Glassman SD, Kantamneni NR, Mugavin MO, Djurasovic M. Clinical outcomes after posterolateral lumbar fusion in workers’ compensation patients: a case-control study. Spine 2010;35: 1812–7. [15] Vaccaro AR, Ring D, Scuderi G, Cohen DS, Garfin SR. Predictors of outcome in patients with chronic back pain and low-grade spondylolisthesis. Spine 1997;22:2030–4; discussion 2035. [16] Choi YH, Mayer T, Williams M, Gatchel RJ. What is the best screening test for depression in chronic spinal pain patients? Spine J 2014;14:1175–82. [17] Spitzer RL, Kroenke K, Williams JB. Validation and utility of a selfreport version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 1999;282:1737–44. [18] Gilbody S, Richards D, Barkham M. Diagnosing depression in primary care using self-completed instruments: UK validation of PHQ-9 and CORE-OM. Br J Gen Pract 2007;57:650–2. [19] Anagnostis C, Gatchel RJ, Mayer TG. The pain disability questionnaire: a new psychometrically sound measure for chronic musculoskeletal disorders. Spine 2004;29:2290–302; discussion 2303. [20] Gatchel RJ, Mayer TG, Theodore BR. The pain disability questionnaire: relationship to one-year functional and psychosocial rehabilitation outcomes. J Occup Rehabil 2006;16:75–94. [21] Wilson H. Minimum clinical important differences of health outcomes in a chronic pain population: are they predictive of poor outcomes?. Arlington, TX: University of Texas at Arlington, 2007. [22] Diagnostic and statistical manual of mental disorders. 5th ed. American Psychiatric Assocation, 2013. [23] L€owe B, Un€utzer J, Callahan CM, Perkins AJ, Kroenke K. Monitoring depression treatment outcomes with the patient health questionnaire-9. Med Care 2004;42:1194–201. [24] Badia X, Diaz-Prieto A, Gorriz MT, Herdman M, Torrado H, Farrero E, et al. Using the EuroQol-5D to measure changes in quality of life 12 months after discharge from an intensive care unit. Intensive Care Med 2001;27:1901–7. [25] Johnson JA, Coons SJ, Ergo A, Szava-Kovats G. Valuation of EuroQOL (EQ-5D) health states in an adult US sample. Pharmacoeconomics 1998;13:421–33. [26] Jansson K-A, Nemeth G, Granath F, J€onsson B, Blomqvist P. Healthrelated quality of life (EQ-5D) before and one year after surgery for lumbar spinal stenosis. J Bone Joint Surg Br 2009;91:210–6. [27] Arpino L, Iavarone A, Parlato C, Moraci A. Prognostic role of depression after lumbar disc surgery. Neurol Sci 2004;25:145–7. [28] Sinikallio S, Aalto T, Airaksinen O, Herno A, Kr€oger H, Viinam€aki H. Depressive burden in the preoperative and early recovery phase predicts poorer surgery outcome among lumbar spinal stenosis patients: a one-year prospective follow-up study. Spine 2009;34:2573–8.
Clinical Study
The impact of preoperative depression on quality of life outcomes after lumbar surgery Jacob A. Miller, BSa,b, Adeeb Derakhshan, BSa,b, Daniel Lubelski, BAa,b, Matthew D. Alvin, MBA, MAa,c, Matthew J. McGirt, MDd, Edward C. Benzel, MDa,b,e, Thomas E. Mroz, MDa,b,e,* a
Cleveland Clinic Center for Spine Health, Cleveland Clinic, 9500 Euclid Ave., S40, Cleveland, OH 44195, USA b Cleveland Clinic Lerner College of Medicine, 9500 Euclid Ave., NA41, Cleveland, OH 44195, USA c Case Western Reserve University School of Medicine, 10900 Euclid Ave., Cleveland, OH 44106, USA d Carolina Neurosurgery and Spine Associates, 225 Baldwin Ave., Charlotte, NC 28204, USA e Department of Neurological Surgery, Cleveland Clinic, 9500 Euclid Ave., S40, Cleveland, OH 44195, USA Received 25 November 2013; revised 13 April 2014; accepted 30 June 2014
Abstract
BACKGROUND CONTEXT: Some, smaller studies have investigated the effect of preoperative depression on postoperative improvement in quality of life (QOL). However, they have not used the Patient Health Questionnaire 9 (PHQ-9) in self-reported depression. PURPOSE: To assess the effect of preoperative depression as measured by the PHQ-9 on postoperative improvement in QOL. STUDY DESIGN: A retrospective review at a single tertiary-care referral center. PATIENT SAMPLE: Patients who underwent lumbar decompression or fusion between 2008 and 2012. OUTCOMES MEASURES: A self-reported EuroQol five-dimensions (EQ-5D) quality-adjusted life-years Index. METHODS: Quality of life data were collected using the institutional prospectively collected database of patient-reported health status measures. The EQ-5D questionnare, PDQ, and PHQ-9 were used. Linear and logistic regression analyses were performed to assess the impact of preoperative depression on QOL improvement. RESULTS: Elevated preoperative pain (PDQ, b50.0017, p5.0009) and worsened depression (PHQ-9, b50.0044, p5.0359) were significantly associated with diminished postoperative improvement in QOL, as measured by the EQ-5D. Furthermore, greater depression (PHQ-9, odds ratio [OR] 0.93, p!.0001) and pain (PDQ, OR 0.99, p5.02) were associated with significantly diminished postoperative improvement exceeding the minimum clinically important difference. CONCLUSIONS: Increased preoperative pain and depression were shown to be associated with significantly reduced improvement in postoperative QOL, as measured by the EQ-5D. Ó 2015 Elsevier Inc. All rights reserved.
Keywords:
Lumbar surgery; EQ-5D; PHQ-9; PDQ; Depression; Psychosocial outcomes
FDA device/drug status: Not applicable. Author disclosures: JAM: Nothing to disclose. AD: Nothing to disclose. DL: Nothing to disclose. MDA: Nothing to disclose. MJM: Grants: Depuy and Stryker Spine (E, paid directly to institution). ECB: Royalties: Elsevier Pub (B), Thieme Pub (B); Stock Ownership: Axiomed (NA), Depuy (NA), Orthomems (NA), Turning Point (NA); Consulting: Axiomed; Speaking and or Teaching Arrangements: Multiple (Varied); Trips/Travel: Multiple (Varied); Grants: Orthopaedic Research and Education Foundation (F, Paid directly to institution), Rawlings (F, Paid directly to institution). TEM: Stock Ownership: PearlDiver, Inc. (No money received); http://dx.doi.org/10.1016/j.spinee.2014.06.020 1529-9430/Ó 2015 Elsevier Inc. All rights reserved.
Consulting: Globus Medical (B); Speaking and/or Teaching Arrangements: AO Spine (B). The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. IRB approval (IRB #11-957) was obtained before initiation of the study. No funding sources or conflict of interests were present. * Corresponding author. Cleveland Clinic Center for Spine Health, Neurological Institute, The Cleveland Clinic, 9500 Euclid Ave., S80, Cleveland, OH 44195, USA. Tel.: (216) 445-9232; fax: (216) 363-2040. E-mail address: [email protected] (T.E. Mroz)
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
59
Introduction In appropriate patients, lumbar decompression and fusion significantly improve the quality of life (QOL) [1–5]. However, such surgical procedures are associated with significant cost and harbor the potential for significant complications [4,6–11]. In an effort to reduce costs and improve the QOL outcomes after spine surgery, researchers have investigated the preoperative risk factors to better identify optimal surgical candidates. These studies have preliminarily identified the risk factors related to comorbidities, demographic, socioeconomic, and psychological factors. Diabetes and smoking, for example, have been shown to increase the incidence of complications and length of stay after lumbar operations for lumbar canal stenosis, prolapsed discs, and thoracolumbar scoliosis [12]. A systematic review of the literature identified a lower level of education, higher preoperative pain level, low work satisfaction, and longer duration of sick leave as risk factors for unfavorable outcomes, as they relate to pain, disability, and return to work [13]. Prior studies have identified a strong correlation between psychological factors, such as depression, and poor clinical outcomes after lumbar spine surgery, as measured by the patient-reported metrics such as the visual analog scale (VAS) and Oswestry Disability Index [4,6–11]. A systematic review of the literature by Celestin et al. [9] noted that preoperative depression, anxiety, and poor coping abilities were associated with reduced surgical benefit after lumbar surgical procedures. Other socioeconomic factors, such as workers’ compensation litigation, have also been implicated in worsened postoperative QOL outcomes [14,15]. These findings support the use of preoperative psychosocial questionnaires to better predict the patients most likely to achieve optimal postoperative QOL outcomes. Although other studies have investigated the effect of preoperative depression on QOL outcomes, none have used the Patient Health Questionnaire 9 (PHQ-9) that has demonstrated the highest sensitivity and specificity in assessing self-reported depression among psychosocial questionnaires [16–18]. Furthermore, no studies have used the Pain Disability Questionnaire (PDQ) in conjunction with depression questionnaires to evaluate the effects of pain or preoperative QOL on postoperative outcomes [19]. In the present study, we sought to assess the impact of preoperative depression, and other baseline patient characteristics, on QOL outcomes after lumbar surgery. We hypothesized that lumbar spine surgery significantly improves the QOL outcomes in all patients, but with a statistically significant lower improvement in those with greater preoperative depression.
Context The negative influence of behavioral health conditions on outcomes after surgical intervention has been surmised on a number of levels. The authors sought to add to this literature by reviewing patient experiences following elective lumbar spine surgery using a variety of quality of life and mental health questionnaires, including the Patient Health Questionnaire-9 (PHQ-9). Contribution The authors evaluated results among a series of 919 patients. Elevated levels of preoperative pain and depression (as measured on the PHQ-9) were found to adversely influence postoperative improvement in quality of life as reflected on the EQ-5D. Implications The authors’ analysis adds to a growing body of literature that improves understanding regarding the influence of behavioral health conditions on outcomes after spine surgery. This study is remarkable due to the amount of data collected on participating patients and the number of individuals included. It remains a retrospective review, however, and one that was performed at a single tertiary center. As a result, the findings may not be generalizable to all individuals encountered in spine surgical practices throughout the nation. —The Editors
between 2008 and 2012, at a single tertiary-care center for nontraumatic and nonneoplastic indications. Quality of life data were collected using the institutional prospectively collected database of patient-reported health status measures, called the Knowledge Program that includes validated questionnaires such as the EuroQol five-dimensions (EQ-5D), PDQ, and PHQ-9, administered at each pre- and postoperative outpatient visit. Patients included in this study were followed for a minimum of 6 months within the study period, with one of the following diagnoses: lumbosacral spondylosis, degenerative disc disease, lumbar spinal stenosis, lumbar disc displacement, lumbar kyphosis, or lumbar scoliosis. Patients were excluded if they were younger than 18 years at the time of surgery, had follow-up of less than 6 months, or had incomplete pre- or postoperative QOL data. QOL outcome measures
Methods Patient selection A retrospective review was performed of all patients who underwent elective lumbar decompression and fusion
Health-status measures questionnaires (PDQ, PHQ-9, and EQ-5D) were collected in a prospective manner, preoperatively, and at each subsequent follow-up visit. The previously validated [19,20] PDQ records functional and psychosocial components of pain, with a total score between 0 and 150, with increasing pain reflected by an
60
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
increasing score. The minimum clinically important difference (MCID) in PDQ has been reported to be 20, although this has not been studied in a lumbar surgery population; rather, it has been studied in a chronic pain population [21]. The PHQ-9 was used for the assessment of depression. This questionnaire assesses for depression based on nine Diagnostic and Statistical Manual of Mental Disorders V [22] criteria for major depressive disorder. Its nine questions are scored from 0 (not at all) to 3 (nearly every day), making the total score range from 0 to 27. The total score is commonly categorized into one of five groups: no symptoms (score50–4), minimal symptoms (score55–9), minor depression (score510–14), moderately severe major depression (score515–19), and severe major depression (score520–27). The diagnostic validity of the PHQ-9 has been established in large multicenter analyses [5,17] and in a chronic spinal pain population [16]. The MCID for the PHQ-9 has been reported to be 5, although this has not been studied in a lumbar surgery population; rather, it has been studied in depressed patients being treated in psychiatric clinics [23]. The EQ-5D contains five dimensions of health state: mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each dimension is scored from 0 to 3. For example, for the anxiety/depression dimension, patients check a box indicating that they are ‘‘not anxious or depressed,’’ ‘‘moderately anxious or depressed,’’ or ‘‘extremely anxious or depressed,’’ which would respectively correlate with increasing scores [24–26]. An EQ5D index score is then calculated from these five dimension subscores. This index score represents the patient’s qualityadjusted life-years (QALYs), ranging from a QALY of 0 (death) to 1 (perfect health), and is included in the analysis as a reflection of the patient’s overall QOL. As such, the EQ-5D index represents a calculated QALY. The MCID for the EQ-5D after surgery has been reported to be 0.1 [4]. Once identified, the patient sample was assessed for the impact of depression (via PHQ-9) on postoperative QOL outcomes.
regression analysis was performed using postoperative EQ-5D QALY change exceeding the MCID of 0.14 as the dependent variable. Again, variables that were associated (p!.2) via simple logistic regression were included in the multiple logistic regression. Regression coefficients (b) were reported for linear regression, and odds ratios (ORs) with 95% confidence intervals were reported for logistic regression. Data were analyzed using the JMP pro 10 statistical software (SAS Institute, Inc., Carey, NC, USA, 2012). For comparisons with multiple t tests, a Bonferroni correction was made, and thus, p values!.01 were considered statistically significant. Otherwise, p values!.05 were considered statistically significant.
Results Demographics Two thousand nine hundred fifty-seven patients were eligible for inclusion. Of these, 1,975 (67%) had a follow-up of less than 6 months and 63 (2%) did not have pre- or postoperative QOL data, and were thus excluded. Hence, 919 patients were included in the study. The cohort had an average follow-up of 421 days. Four hundred sixty-eight (50.9%) patients were men and 833 (90.6%) were Caucasian (Table 1). The average body mass index was 29.1, Table 1 Baseline presentation of 919 patients who underwent lumbar surgery Baseline
n5919
Age (y) Male gender BMI Race
59.6613.7 468 (50.9%) 29.165.8 Asian: 2 (0.2%) Black: 58 (6.3%) Caucasian: 833 (90.6%) Mixed: 4 (0.5%) Other/unknown: 22 (2.4%) Renal failure: 20 (2.2%) Diabetes: 153 (16.7%) Coronary artery disease: 78 (8.5%) Hypertension: 356 (38.8%) Stroke history: 17 (1.9%) $53,399 272 (29.6%) Lumbosacral spondylosis: 118 (12.8%) Degenerative disc disease: 84 (9.1%) Spinal stenosis: 91 (9.9%) Disc displacement: 515 (56.2%) Lumbar kyphosis: 28 (3.0%) Lumbar scoliosis: 83 (9.0%) Laminotomy: 239 (26.0%) Laminoplasty: 15 (1.6%) Laminectomy: 174 (18.9%) Anterior fusion: 12 (1.3%) Posterolateral fusion: 267 (29.1%) Posterior fusion: 212 (23.1%) 421.26216.7 179 (19.5%)
Comorbidities
Statistical analysis The primary aim of this study was to identify if there was an association between preoperative depression (as assessed by PHQ-9) and postoperative improvement in EQ5D QALY scores. Continuous data were expressed as mean6standard deviation and compared via the Student t test; paired data (eg, preoperative and postoperative QALYs) were compared with a paired t test. Categorical data were compared with the Fisher Exact test. To avoid overfitting the model, only variables that demonstrated association (p!.2) with improvements in EQ-5D QALY (via simple linear regression) were analyzed via a multiple linear regression model to assess for independent predictors of improved QOL outcomes. Preoperative EQ-5D QALY was included in the model to adjust for the ceiling effect and normalize all improvements to baseline scores. A logistic
Income (median) Antidepressants Diagnosis
Surgery type
Follow-up (d) Depressed (PHQ-9O14)
BMI, body mass index; PHQ-9, patient health questionnaire 9.
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
and the median annual income was $53,399. A significant portion of the population had comorbidities at the time of surgery: 20 (2.2%) had chronic renal failure, 153 (16.7%) had diabetes, 78 (8.5%) had coronary artery disease, 356 (38.8%) had hypertension, and 17 (1.9%) had a history of cerebral infarction. There were 272 (29.6%) patients taking antidepressants and 179 (19.5%) reporting moderate or severe depression (PHQ-9 score greater than 14). Four hundred ninety-one (53.5%) patients underwent lumbar fusion procedures, whereas the remaining 428 (46.5%) underwent lumbar decompression without fusion. Anterior interbody fusions accounted for 1.3% of the study population, whereas posterolateral and posterior interbody fusions accounted for 29.1% and 23.1%, respectively. The indications for these procedures varied, although more than half (56.2%) were for diagnoses of disc displacement. Postoperative health-related QOL outcomes At the last follow-up, statistically significant improvements in QOL outcomes in all questionnaires were observed in the cohort (Table 2). The following pre- and postoperative averages were observed: PDQ functional component (55.2 preoperatively vs. 34.4 postoperatively), PDQ psychosocial component (31.0 preoperatively vs. 19.1 postoperatively), PDQ total (86.3 preoperatively vs. 53.4 postoperatively), PHQ-9 (9.8 preoperatively vs. 5.1 postoperatively), and EQ-5D QALY (0.49 preoperatively Table 2 Pre- and postoperative quality of life outcomes Quality of life measure PDQ FC Preoperative Postopoperative Change PDQ PC Preoperative Postoperative Change PDQ total Preoperative Postoperative Change PHQ-9 Preoperative Postoperative Change EQ-5D index Preoperative Postoperative Change EQ-5D QALYOMCID (0.1) PHQ-9OMCID (5) PDQOMCID (20)
Lumbar surgeries (n5919)
p
55.2618.0 34.4625.7 20.3625.1
!.0001
31.0613.1 19.1616.6 11.7614.9
!.0001
86.3629.4 53.4641.2 32.2638.5
!.0001
9.866.8 5.165.6 3.566.3
!.0001
0.4960.22 0.6560.22 0.1560.25 461 (50.2%) 347 (37.8%) 532 (57.9%)
!.0001
— — —
EQ-5D, EuroQol five-dimensions; PDQ, pain disability questionnaire; FC, functional component; PC, psychosocial component; PHQ-9, patient health questionnaire 9; QALY, quality-adjusted life-years; MCID, minimum clinically important difference. Note: Paired Student t test was used to check for significance (p!.01).
61
vs. 0.65 postoperatively) (p!.0001 for all questionnaires). Four hundred sixty-one (50.2%) patients experienced improvement exceeding the MCID of 0.1 in EQ-5D QALY. Additionally, 347 (37.9%) and 532 (57.9%) patients experienced improvement exceeding the MCID of 5 in PHQ-9 and of 20 in PDQ, respectively. Regression analysis Simple and multiple linear regression analyses were performed to assess the effect of preoperative depression, as measured by the PHQ-9 on postoperative QOL improvement (Table 3). For both of these analyses, EQ-5D QALY improvement (postoperative minus preoperative QALYs) was used as the dependent outcome variable. Simple linear regression revealed six variables that were moderately associated (p!.2) with improvement: age (b50.0006, p5.1866), procedures with fusion (b50.0356, p5.0032), procedure type (b50.6058, p5.0050), preoperative EQ5D QALY (b50.6389, p!.0001), preoperative PHQ-9 score (b50.0083, p!.0001), and preoperative PDQ score (b50.0021, p!.0001). Both ‘‘procedure type’’ and ‘‘procedures with fusion’’ were included as separate variables in the model to test for the independent effect of fusion. These six variables were included in a multiple linear regression model. After modeling, age (b50.0008, p5.2315), procedure type (b50.0612, p5.79), and lumbar fusion (b50.0006, p5.6344) were no longer significant predictors of EQ-5D QALY improvement. The remaining three variables (preoperative EQ-5D QALY, PHQ-9, and PDQ scores) Table 3 Linear regression analysis of independent preoperative predictors of improvement in health state (change in EQ-5D score) after lumbar surgery Simple regression
Multiple regression
Variables
Coefficient
p
Coefficient
Age (y) Race* Female gender Income BMI Lumbar fusion Procedure typey Diagnostic indicationz Renal failure Diabetes Coronary artery disease Hypertension Stroke history Antidepressant use EQ-5D QALY baseline PHQ-9 baseline PDQ baseline
0.0006 0.0018 0.0048 !0.0001 0.0001 0.0356 0.6058 0.0017 0.0094 0.0002 0.0096 0.0028 0.0206 0.0025 0.6389 0.0083 0.0021
.1866 .8983 .5652 .2825 .7642 .0032 .0050 .5501 .7380 .9865 .5185 .7423 .5014 .7836 !.0001 !.0001 !.0001
p
0.0008
.2315
0.0006 0.0612
.6344 .7906
0.8915 0.0044 0.0017
!.001 .0359 .0009
EQ-5D, EuroQol five-dimensions; QALY, quality-adjusted life-years; PHQ-9, patient health questionnaire 9; PDQ, pain disability questionnaire; BMI, body mass index. * Coefficient reported for Caucasian. y Coefficient reported for laminectomy. z Coefficient reported for disc displacement.
62
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
were all found to be significant independent predictors of EQ-5D QALY improvement. The regression coefficients were all negative, indicating that increasing preoperative pain (PDQ, b50.0017, p5.0009) and worsening PHQ-9 depression index (PHQ-9, b50.0044, p!.0359) reduced postoperative improvement. The normalization variable, EQ-5D QALY, was also a significant predictor in the model (b50.8915, p!.001). Simple and multiple logistic regression analyses were also performed to assess the effect of preoperative PHQ-9 depression index on postoperative QOL improvement exceeding the MCID (Table 4). For both of these analyses, EQ-5D QALY improvement exceeding the MCID (0.1) was used as the dependent outcome variable. Simple logistic regression identified seven variables that were moderately associated (p!.2) with EQ-5D QALY exceeding the MCID: performance of lumbar fusion (OR 1.62; 95% confidence intervals: [1.10, 2.38], p5.01), procedure type (OR 0.20; [0.15, 0.80], p5.002), diagnostic indication (OR 0.96; [0.92, 1.01], p5.10), history of stroke (OR 2.44; [0.90, 7.73], p5.08), preoperative EQ-5D QALY (OR 0.005; [0.002, 0.01], p!.0001), preoperative PHQ-9 (OR 1.06; [1.04, 1.09], p!.0001), and preoperative PDQ (OR 1.02; [1.01, 1.03], p!.0001). These seven variables were included in a multiple logistic regression. After modeling, only preoperative EQ-5D QALY (OR 0.001; [0.0003, 0.002], p!.0001), PHQ-9 (OR 0.93; [0.90, 0.96], p!.0001), and PDQ (OR 0.99; [0.97, 0.99], p5.02) remained independent significant predictors of QALY change exceeding the MCID. The ORs for these three variables were all less than 1, indicating that worsening PHQ-9 depression index
(PHQ-9, OR 0.93, p!.0001) and pain (PDQ, OR 0.99, p5.02) reduced postoperative improvement exceeding the MCID. The normalization variable, EQ-5D QALY, was also a significant predictor in the model (OR 0.001, p!.0001).
Discussion In the present study, we sought to investigate the effect of preoperative depression as measured by the PHQ-9 on postoperative QOL through the use of three patient reported questionnaires (PDQ, PHQ-9, and EQ-5D). Multiple linear regression showed that each of these health status measures was, individually, a significant predictor of improvement in EQ-5D QALY, with greater PHQ-9 depression index and pain being predictive of worse postoperative QOL outcomes. Although the present study is the first to use the PDQ and PHQ-9 in postoperative QOL assessment after lumbar surgery, several studies have used other tools in assessing the association between depression and outcome. Arpino et al. [27] used the Zung depression scale in their prospective investigation of 73 patients who underwent microdiscectomy for lumbar disc herniation. The authors concluded that patients with lower preoperative Zung depression scores (less depression) had a better outcome in pain (as measured by VAS) at 3 and 12 months after surgery. In a prospective clinical study, Sinikallio et al. [28] used the Beck Depression Inventory (BDI) to assess the 102 patients who underwent decompression for lumbar spinal stenosis. They reported that higher preoperative BDI scores were independently
Table 4 Logistic regression analysis of independent preoperative predictors of MCID Simple regression
Multiple regression
Variables
OR (95% CI)
p
Age (y) Race* Female gender Income BMI Lumbar fusion Procedure typey Diagnostic indicationz Renal failure Diabetes Coronary artery disease Hypertension Stroke history Antidepressant use EQ-5D QALY baseline PHQ-9 baseline PDQ baseline
1.57 1.33 1.07 1.01 0.98 1.62 0.20 0.96 0.66 0.90 0.82 0.98 2.44 1.16 0.005 1.06 1.02
.20 .20 .62 .98 .50 .01 .002 .10 .37 .54 .41 .86 .08 .30 !.0001 !.0001 !.0001
(0.78, 3.15) (0.86, 2.09) (0.82, 1.38) (0.43, 0.99) (0.93, 1.04) (1.10, 2.38) (0.15, 0.80) (0.92, 1.01) (0.26, 1.61) (0.63, 1.27) (0.51, 1.31) (0.75, 1.28) (0.90, 7.73) (0.87, 1.55) (0.002, 0.01) (1.04, 1.09) (1.01, 1.03)
OR (95% CI)
p
1.08 (0.52, 2.30) 0.80 (0.56, 1.25) 0.97 (0.90, 1.04)
.83 .65 .36
1.00 (0.99, 1.01)
1.00
0.001 (0.0003, 0.002) 0.93 (0.90, 0.96) 0.99 (0.97, 0.99)
!.0001 !.0001 .02
MCID, minimum clinically important difference; OR, odds ratio; CI, confidence interval; BMI, body mass index; EQ-5D, EuroQol five-dimensions; QALY, quality-adjusted life-years; PHQ-9, patient health questionnaire 9; PDQ, pain disability questionnaire. * Coefficient reported for Caucasian. y Coefficient reported for laminectomy. z Coefficient reported for disc displacement.
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
predictive of poorer self-reported functional ability, symptom severity, and poorer walking capacity at 1-year follow-up. Chaichana et al. [8] used the Zung self-rating depression scale in their prospective study of 67 patients who underwent lumbar discectomy. They determined that a patient with higher preoperative depression was significantly less likely to achieve an MCID in disability or QOL after surgery, thereby reaffirming the predictive value of preoperative depression in postoperative outcome. Adogwa et al. [10] used the Zung depression scale to retrospectively review the outcomes of 150 patients who underwent revision lumbar surgery for symptomatic pseudarthrosis, adjacent segment disease, and same-level recurrent stenosis. They concluded that higher (worse) preoperative Zung depression scores were independently associated with a decreased 2year improvement in QOL outcome in these patients. In a separate study, Adogwa et al. [11] retrospectively investigated 53 patients who underwent revision surgery for same-level lumbar stenosis. Using the Zung self-rating depression scale, the authors evaluated the QOL outcomes (via the EQ-5D, Oswestry Disability Index, and the pain VAS) 2 years after surgery and concluded that increasing (worsening) preoperative Zung depression score was statistically and clinically significantly associated with diminished improvement in QOL outcome postoperatively. Our results corroborate the results of these studies and expand the patient population to include individuals with all types of lumbar decompression or fusion. The studies referenced in the previous paragraph did not use the PHQ-9 measure to assess the impact of depression on postoperative QOL outcomes. The purpose of this study was to, therefore, investigate the effect of preoperative depression using the PHQ-9 on QOL outcomes after decompression and fusion lumbar surgeries in a large cohort. Our initial hypothesis that greater preoperative PHQ-9 depression index would lead to decreased postoperative improvement was confirmed through analysis via multiple linear and logistic regressions. This suggests that the PHQ-9 is at least on par with the Zung self-rating depression scale and the BDI. Following statistical analysis, several significant results were noted. There were significant differences between the preoperative and postoperative PDQ (both the functional and psychosocial components), PHQ-9, and EQ-5D index (p!.0001 for all). This suggests, as expected, that surgery improved the patients’ QOL in general. In the multiple linear regression analysis, the baseline EQ-5D QALY score (p!.001), baseline PHQ-9 score (p5.0359), and baseline PDQ score (p5.0009) were all independently and statistically significant predictive of an improvement in health state (as measured by the change in EQ-5D QALY) after lumbar surgery. Additionally, these variables were also significant predictors of change in EQ-5D QALY exceeding the MCID, demonstrating that these results have clinical and statistical significance. It is important to include the EQ-5D QALY baseline score in the model, as it corrects for all levels of preoperative QOL, and therefore, adjusts for the ceiling
63
effect. Furthermore, inclusion of the variable generalizes the model to include patients with any preoperative EQ5D QALY, whether low or high. Notably, potential confounding factors such as diagnostic indication for surgery were not independently associated with improvement. This study includes the largest cohort of lumbar spine surgery patients to date for the assessment of variables associated with changes in QOL. We assessed for depression using the PHQ-9, based on the Diagnostic and Statistical Manual of Mental Disorders V criteria. This study contributes to the body of evidence that demonstrates the negative impact of preoperative depression on postoperative QOL outcomes. Furthermore, the model is generalizable to surgeries with and without fusion. Previous smaller studies support our findings, although these studies used QOL surveys with reduced sensitivity and specificity in detecting depression [6– 9,11,15,16,27,28]. Although 919 patients were included after exclusion, a significant number of patients (1,975) were excluded because of follow-up of less than 6 months. This loss to follow-up is a potential confounding factor in the data: individuals with a follow-up of less than 6 months may have superior postoperative outcomes independent of depression state, and thus, the observed effects in the present study may pertain only to those with long follow-up times, given the chronic nature of their disease. Furthermore, it is more likely that individuals with long follow-up times are depressed, given the length of time they experience symptoms. The results of this study have significant clinical implications. This study suggests that depression is a modifiable risk factor for poor outcomes. Surgeons should be cognizant of the fact that patients that are preoperatively depressed are more likely to experience poorer outcomes after lumbar surgery. As such, greater attention should be paid to accurately evaluate all patients preoperatively with a validated psychosocial questionnaire. A trained multidisciplinary team may then address and treat depression before progressing to surgical intervention in an effort to maximize the benefit of surgery. Finally, depression should be addressed in the postoperative follow-up in an effort to prevent relapse into depression, as it may diminish the postoperative benefits of surgery. The results of this study should be interpreted with consideration of the study’s limitations. Because of the perceived stigma of depression, patients may not honestly and objectively report their depression. Moreover, the patient’s emotional status may vary between visits and, as such, answers given on the questionnaire may not be an accurate portrayal of how the patient typically feels. Additionally, this was a retrospective study with a relatively short follow-up period of 1 year. Prospectively designed studies with longer follow-ups are needed to further validate the findings. Finally, as this study was conducted at a single tertiary-care institution, the external validity may be limited to institutions with similar patient populations and demographics. Despite these limitations, the methods and assumptions in the present study adhere similarly to
64
J.A. Miller et al. / The Spine Journal 15 (2015) 58–64
what has been published previously in the literature, thereby allowing us to justify our conclusions.
Conclusion Worse preoperative pain and depression are significantly associated with diminished improvement in QOL after lumbar surgery. These findings support the use of preoperative psychosocial questionnaires that may serve as a tool for selecting patients more likely to have superior postoperative outcomes and those who are unlikely to derive benefit from the surgery. Patients should be made aware of these associations when deciding on surgical treatment. References [1] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Aaronson O, Cheng JS, et al. Cost per quality-adjusted life year gained of revision neural decompression and instrumented fusion for same-level recurrent lumbar stenosis: defining the value of surgical intervention. J Neurosurg Spine 2012;16:135–40. [2] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Devin CJ, Cheng JS, et al. Cost per quality-adjusted life year gained of laminectomy and extension of instrumented fusion for adjacent-segment disease: defining the value of surgical intervention. J Neurosurg Spine 2012;16:141–6. [3] Adogwa O, Parker SL, Davis BJ, Aaronson O, Devin C, Cheng JS, et al. Cost-effectiveness of transforaminal lumbar interbody fusion for Grade I degenerative spondylolisthesis. J Neurosurg Spine 2011;15:138–43. [4] Parker SL, Adogwa O, Mendenhall SK, Shau DN, Anderson WN, Cheng JS, et al. Determination of minimum clinically important difference (MCID) in pain, disability, and quality of life after revision fusion for symptomatic pseudoarthrosis. Spine J 2012;12:1122–8. [5] Tharin S, Mayer E, Krishnaney A. Lumbar microdiscectomy and lumbar decompression improve functional outcomes and depression scores. Evid Based Spine Care J 2012;3:65–6. [6] Trief PM, Grant W, Fredrickson B. A prospective study of psychological predictors of lumbar surgery outcome. Spine 2000;25:2616–21. [7] LaCaille RA, DeBerard MS, Masters KS, Colledge AL, Bacon W. Presurgical biopsychosocial factors predict multidimensional patient: outcomes of interbody cage lumbar fusion. Spine J 2005;5:71–8. [8] Chaichana KL, Mukherjee D, Adogwa O, Cheng JS, McGirt MJ. Correlation of preoperative depression and somatic perception scales with postoperative disability and quality of life after lumbar discectomy. J Neurosurg Spine 2011;14:261–7. [9] Celestin J, Edwards RR, Jamison RN. Pretreatment psychosocial variables as predictors of outcomes following lumbar surgery and spinal cord stimulation: a systematic review and literature synthesis. Pain Med 2009;10:639–53. [10] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Aaronson OS, Cheng JS, et al. Preoperative Zung depression scale predicts outcome after revision lumbar surgery for adjacent segment disease, recurrent stenosis, and pseudarthrosis. Spine J 2012;12:179–85.
[11] Adogwa O, Parker SL, Shau DN, Mendenhall SK, Bydon A, Cheng JS, et al. Preoperative Zung depression scale predicts patient satisfaction independent of the extent of improvement after revision lumbar surgery. Spine J 2013;13:501–6. [12] Appaduray SP, Lo P. Effects of diabetes and smoking on lumbar spinal surgery outcomes. J Clin Neurosci 2013;20:1713–7. [13] den Boer JJ, Oostendorp RAB, Beems T, Munneke M, Oerlemans M, Evers AWM. A systematic review of bio-psychosocial risk factors for an unfavourable outcome after lumbar disc surgery. Eur Spine J 2006;15:527–36. [14] Carreon LY, Glassman SD, Kantamneni NR, Mugavin MO, Djurasovic M. Clinical outcomes after posterolateral lumbar fusion in workers’ compensation patients: a case-control study. Spine 2010;35: 1812–7. [15] Vaccaro AR, Ring D, Scuderi G, Cohen DS, Garfin SR. Predictors of outcome in patients with chronic back pain and low-grade spondylolisthesis. Spine 1997;22:2030–4; discussion 2035. [16] Choi YH, Mayer T, Williams M, Gatchel RJ. What is the best screening test for depression in chronic spinal pain patients? Spine J 2014;14:1175–82. [17] Spitzer RL, Kroenke K, Williams JB. Validation and utility of a selfreport version of PRIME-MD: the PHQ primary care study. Primary Care Evaluation of Mental Disorders. Patient Health Questionnaire. JAMA 1999;282:1737–44. [18] Gilbody S, Richards D, Barkham M. Diagnosing depression in primary care using self-completed instruments: UK validation of PHQ-9 and CORE-OM. Br J Gen Pract 2007;57:650–2. [19] Anagnostis C, Gatchel RJ, Mayer TG. The pain disability questionnaire: a new psychometrically sound measure for chronic musculoskeletal disorders. Spine 2004;29:2290–302; discussion 2303. [20] Gatchel RJ, Mayer TG, Theodore BR. The pain disability questionnaire: relationship to one-year functional and psychosocial rehabilitation outcomes. J Occup Rehabil 2006;16:75–94. [21] Wilson H. Minimum clinical important differences of health outcomes in a chronic pain population: are they predictive of poor outcomes?. Arlington, TX: University of Texas at Arlington, 2007. [22] Diagnostic and statistical manual of mental disorders. 5th ed. American Psychiatric Assocation, 2013. [23] L€owe B, Un€utzer J, Callahan CM, Perkins AJ, Kroenke K. Monitoring depression treatment outcomes with the patient health questionnaire-9. Med Care 2004;42:1194–201. [24] Badia X, Diaz-Prieto A, Gorriz MT, Herdman M, Torrado H, Farrero E, et al. Using the EuroQol-5D to measure changes in quality of life 12 months after discharge from an intensive care unit. Intensive Care Med 2001;27:1901–7. [25] Johnson JA, Coons SJ, Ergo A, Szava-Kovats G. Valuation of EuroQOL (EQ-5D) health states in an adult US sample. Pharmacoeconomics 1998;13:421–33. [26] Jansson K-A, Nemeth G, Granath F, J€onsson B, Blomqvist P. Healthrelated quality of life (EQ-5D) before and one year after surgery for lumbar spinal stenosis. J Bone Joint Surg Br 2009;91:210–6. [27] Arpino L, Iavarone A, Parlato C, Moraci A. Prognostic role of depression after lumbar disc surgery. Neurol Sci 2004;25:145–7. [28] Sinikallio S, Aalto T, Airaksinen O, Herno A, Kr€oger H, Viinam€aki H. Depressive burden in the preoperative and early recovery phase predicts poorer surgery outcome among lumbar spinal stenosis patients: a one-year prospective follow-up study. Spine 2009;34:2573–8.