Preoperative Depression, Lumbar Fusion, and Opioid Use

An Assessment of Postoperative Prescription, Quality, and Economic Outcomes

Chloe O'Connell, BS; Tej Deepak Azad, MS; Vaishali Mittal, BS; Daniel Vail, BA; Eli Johnson, BS; Atman Desai, MD; Eric Sun, MD, PhD; John K. Ratliff, MD; Anand Veeravagu, MD


Neurosurg Focus. 2018;44(1):e5 

In This Article


In the present study, we sought to examine the effects of a preoperative diagnosis of depression on a variety of functional outcomes following lumbar fusion surgery using a nationwide administrative sample. We found that a diagnosis of depression prior to lumbar fusion was associated with increased postoperative opioid use, even after controlling for preoperative opioid use, demographic factors, and relevant comorbidities. In addition, preoperative depression was associated with increased rates of complications, revisions, and 30-day readmissions, as well as increased 1- and 2-year costs. Depression diagnosis was not significantly associated with the likelihood of discharge to home after lumbar fusion surgery.

While previous research has implicated preoperative depression in adverse outcomes of lumbar fusion surgery,[4,6,24] our study is the first, to our knowledge, to investigate this topic using a nationwide administrative database containing records from privately insured individuals and patients with Medicare. The consistent nature of this relationship between depression and adverse outcomes supports the notion of depression as a disease affecting both physical and mental health as well as health care utilization. The association between depression and both 1- and 2-year costs highlights the fact that the economic burden of depression, while well established in the general population,[13] plays an important role in the long-term costs after lumbar fusion. While the administrative nature of data in the current study precluded us from assessing the treatment status of patients prior to surgery, in the future it would be worthwhile to determine whether preoperative treatment of depression would improve outcomes. Given the high prevalence of spondylolisthesis in this cohort that underwent lumbar fusion, future research investigating the specific effect of depression diagnosis within this subgroup would be of great utility.

Opioid use is common following lumbar fusion, and the rising rate of opioid abuse in recent years has become a growing problem in this country.[30] After controlling for demographic factors and comorbidities, we found that a preoperative depression diagnosis was associated with increased cumulative MMEs and risk of chronic use 3–12 months after surgery, as well as a decreased probability of opioid cessation 6–12 months postoperatively. Previous studies have identified depression among the numerous factors increasing the risk of opioid use after surgeries such as total hip arthroplasty and bariatric surgery;[9,27] however, its role in lumbar fusion has remained unexplored. Our results suggest that depression may be an important driver of postoperative opioid use in this surgical population, even after controlling for preoperative use. This relationship between depression and postoperative opioid use, particularly in patients with a diagnosis of spondylolisthesis, warrants further investigation.

Study Limitations

Results of the present study must be interpreted considering the limitations inherent to all large database studies. Specifically, our categorization of depression diagnosis, along with comorbidities and outcomes, assumed the accuracy of our administrative data. Although we used a definition of depression that was based on a previously validated report, the incidence of depression in administrative data based on ICD-9 codes may underestimate the true prevalence.[12] Additionally, depression is commonly thought to be a spectrum of disorders rather than one distinct entity; therefore, characterizing it with one group of ICD-9 codes may not achieve the granularity necessary to distinguish between different forms of the disorder.

While generalized anxiety or panic disorder was included as a covariate in our analysis, other psychiatric conditions, such as Axis II disorders and other mood disorders, particularly bipolar disorder and other anxiety disorders, are frequently comorbid with depression and may affect the relationship between depression and outcomes.[20] Additionally, we did not control for health care utilization, which has been associated with a diagnosis of depression as well.[10] Further work is required to investigate the nature of this relationship and its effect on the association between depression and lumbar fusion outcomes.

Finally, opioid use was assessed based on the number and dosage of medications that were both prescribed to the patient and picked up at a pharmacy, not on actual patient ingestion, which may not reflect opioids obtained from other sources not covered by insurance. To more definitively characterize the relationship between depression and opioid use, future studies using more granular data on both depression diagnosis and opioid use would be of great value.