With ~30 million patients diagnosed, diabetes mellitus is one of the most prevalent medical conditions in the United States.1 The correlation of type 2 diabetes with poor outcomes after orthopedic surgery has been well-established in the literature. In regards to orthopedic surgery, diabetes is associated with a spectrum of postoperative complications ranging from delayed wound healing to cardiovascular events to urinary tract infections.2–5 In relation, the presence of diabetes as comorbidity is associated with increased length of hospital stay and greater overall hospital expenditures compared with individuals without the condition.2,6–8
Spine surgery is no different from other orthopedic procedures, with diabetes being associated with an increased risk for acute postoperative complications, longer inpatient length of stay, and higher total hospital costs compared with nondiabetics.7,9–13 However, very few studies have investigated the effects of diabetes on the postoperative course after minimally invasive procedures, such as minimally invasive transforaminal lumbar interbody fusion (MIS TLIF). As such, the purpose of this study is to determine if the presence of diabetes mellitus as comorbidity is associated with complications, inpatient length of stay, and direct hospital costs following MIS TLIF.
A prospectively maintained surgical registry of patients undergoing primary, single-level MIS TLIF for degenerative pathology between 2008 and 2016 was retrospectively reviewed. All patients were treated by a single surgeon at a single academic institution. Patients who underwent MIS TLIF for nondegenerative pathology, such as trauma or infection, or those with incomplete direct hospital cost data were excluded from this analysis.
A unilateral Wiltse approach was performed under fluoroscopic guidance through a paramedian skin incision. The careful precaution was taken to preserve muscular structures and midline ligaments. The transforaminal disk space was visualized through a 21-mm nonexpendable tubular retractor. An ipsilateral laminectomy, facetectomy, and bilateral decompression were completed. The traversing nerve roots were determined and protected. The disk space was prepared using curettes, rongeurs, and rasps to provide space for interbody cage placement. The intervertebral cage was filled with local bone graft taken from facetectomy and laminectomy and the iliac crest. In a minority of cases, bone morphogenetic protein-2 and other bone graft substitutes were also utilized. In addition, unilateral pedicle screws were placed ipsilateral to the approach.
Patients were stratified into either diabetic or nondiabetic cohorts, and propensity matched in a 1:1 manner for age, sex, and comorbidity burden. Preoperative demographic variables included age, sex, body mass index, smoking status, and Charlson Comorbidity Index (CCI). The matching variable comorbidity burden was evaluated by the patient CCI score. Perioperative variables included operative time, estimated intraoperative blood loss, length of stay, time to discharge, and complications. Postoperative complications were identified as altered mental status, arrhythmias, pneumonia, pulmonary embolism, deep vein thrombosis, acute renal failure, urinary retention, urinary tract infection, anemia, epidural hematoma formation, wound dehiscence, and surgical site infection. We conducted a review of hospital financial records and recorded direct hospital costs for each medical treatment category. Each patient’s operative encounter and perioperative length of stay was identified, and costs were summarized by category, including blood collection services, cardiology, intensive care unit, laboratory, neurology, nursing, pharmacy, physical therapy/occupational therapy/speech therapy, radiology, surgical services, and other services.
Statistical analysis was performed using Stata/MP 13.1 for Mac (StataCorp LP, College Station, TX). The association between diabetic status and demographic or perioperative outcomes was tested using Student t test or χ2 analysis for continuous or categorical variables, respectively. Multivariate linear regression was used to test for an association between diabetic status and direct hospital costs. Statistical significance was set at P<0.05.
After 1:1 propensity matching, a total of 50 diabetics and 50 nondiabetics were included in this analysis. Table 1 exhibits demographic data for diabetics and nondiabetics undergoing MIS TLIF. The average age of patients was 58.7 years, with 70.0% (70) male individuals. The average preoperative hemoglobin A1c in the diabetic cohort was 7.4 (range, 5.2–11.2; SD: 1.4) with 16 patients classified as insulin dependent. There were no significant differences in age, sex, body mass index, smoking status, or CCI between propensity-matched patients with and without diabetes (P≥0.05 for each).
Table 2 represents perioperative data for diabetics and nondiabetics undergoing MIS TLIF. No significant differences existed in operative time or intraoperative blood loss (P≥0.05 for each). The postoperative complication rate for diabetics was 10% (5) and for nondiabetics 14% (7) (P=0.552). Postoperative complications included: urinary retention requiring urology consult and discharged with catheterization (2, diabetic; 2, nondiabetic); epidural hematoma confirmed with computerized tomography or magnetic resonance imaging (2, nondiabetic); altered mental status resulting in transfer to neurology intensive care unit (ICU) (1, diabetic); acute respiratory distress requiring intubation (1, nondiabetic); acute renal failure resulting in a transfer to medicine ICU (1, nondiabetic); arrhythmia requiring cardiology consult (1, nondiabetic); wound dehiscence (1, diabetic); and surgical site infection (1, diabetic). In regards to the length of stay, no significant differences existed between diabetic and nondiabetic groups (68.7 vs. 58.3 h, P=0.218). In addition, there was no difference in the time to discharge for diabetics compared with nondiabetics (0–48 h: 30% vs. 38%; 48–72 h: 32% vs. 34%; 72–96 h: 18% vs. 16%; 96–120 h: 12% vs. 8%; over 120 h: 4% vs. 2%; P=0.574) (Fig. 1).
Table 3 compares direct hospital costs between diabetic and nondiabetic patients. Multivariate analysis indicated that diabetes was not associated with differences in total direct hospital costs (US$20,428 vs. US$20,429, P=0.792). In addition, there was no difference between diabetics and nondiabetics in the cost subcategory analysis including charges from various subspecialties, radiology, laboratory, nursing, intensive care unit, pharmacy, therapy, surgery, or other services after MIS TLIF (P≥0.05 for each).
In spine surgery, the presence of diabetes as comorbidity has been associated with an increased risk for acute postoperative complications, length of inpatient stay, and hospital costs.3,4,9–11,14,15 However, few studies have investigated the effect of diabetes on postoperative outcomes in procedure-specific populations, especially minimally invasive lumbar fusion. As such, the purpose of this study is to determine if the presence of diabetes mellitus as comorbidity is associated with complication rates, inpatient length of stay, and direct hospital costs after primary, single-level MIS TLIF.
In this investigation, there was no difference in length of stay between diabetic and nondiabetic patients undergoing MIS TLIF. However, the literature suggests diabetes is a risk factor for an extended hospital stay. Golinvaux and colleagues retrospectively evaluated lumbar fusion patients on the National Surgical Quality Improvement Program (NSQIP) database. Defined as a hospital stay longer than 5 days, patients with noninsulin-dependent diabetes (risk ratio, 1.2; P=0.003) and insulin-dependent diabetes (risk ratio, 1.5, P<0.001) were more likely to have a greater length of stay than nondiabetics.1 Furthermore, Guzman et al3 evaluated patients undergoing lumbar spine surgery for degenerative disease and demonstrated diabetes was a risk factor for an extended inpatient stay (uncontrolled diabetes: 6.02 d; controlled diabetes: 4.07 d; no diabetes: 3.65 d; P<0.001). Previous studies suggest that diabetes may be a risk factor for more extreme lengths of stay after lumbar fusion, however, this risk may not be as evident for minimally invasive approaches that typically require a much shorter hospital course.16
In our study, there was no statistical difference in postoperative complication rates between diabetic and nondiabetic patients undergoing MIS TLIF. However, the literature indicates that diabetes is associated with a higher complication rate in lumbar fusion patients. In a retrospective review of patients who underwent lumbar fusions, Guzman et al3 determined patients with diabetes are more likely to experience acute complications including cerebrovascular [odds ratio (OR), 1.56; P=0.001], respiratory (OR, 1.23; P<0.001), cardiac (OR, 1.41; P<0.001), deep vein thrombosis (OR, 1.5; P<0.001), genitourinary (OR, 1.35; P<0.001), postoperative infection (OR, 1.41; P<0.001), postoperative hemorrhage (OR, 1.31; P<0.001) and overall mortality (OR, 1.44; P=0.001) compared with patients without diabetes. The differences in postoperative complications regarding diabetics between minimally invasive and open lumbar fusions may be related to a shorter hospital stay, which in turn reduces the chance of developing direct and indirect nosocomial morbidity for patients who underwent MIS TLIF.
In our study, there was no difference in total hospital costs between diabetic and nondiabetic patients undergoing MIS TLIF. However, the literature suggests diabetes is associated with greater hospital costs in traditional lumbar fusions. Browne et al12 used the National Inpatient Sample (NIS) database to retrospectively analyze lumbar fusion patients and determined that diabetics had increased total hospital charges (US$39,371 vs. US$38,071; P<0.001). In a similar study using the NIS database, Guzman et al3 investigated patients undergoing surgery for lumbar degenerative disease and determined that uncontrolled diabetics have significantly higher hospital costs than patients without diabetes (US$26,476 vs. US$21,250; P<0.001). On average, the higher costs incurred by diabetics in traditional lumbar fusions are probably explained by longer stays and more difficult postoperative courses. However, increased hospital costs are moderated by MIS TLIF because of a less extensive technique, leading to shorter hospital stays and avoidance of hospital-related complications.
Our study demonstrates that MIS TLIF is effective in diabetic patients with no additional risk compared with nondiabetic patients. Open TLIF is a comparable procedure that was not included in our study. Therefore, we were unable to determine any differences in performance between MIS and open techniques for diabetic patients. Through a less extensive approach compared with open, MIS TLIF has allowed for faster recovery and avoidance of nosocomial complications, leading to improved postoperative outcomes and diminished economic burden on patients.5 Compared with open TLIF procedures, Singh et al demonstrate that MIS TLIF has a shorter anesthesia time (180.6 vs. 258.6 min; P<0.001) and estimated blood loss (124.4 vs. 380.3 mL; P<0.001). Furthermore, patients who underwent MIS TLIF had a shorter length of stay than open TLIF (2.3 vs. 2.9 d; P=0.018) along with lower total hospital costs (US$19,512 vs. US$23,550, P<0.001).16 Likewise, in their meta-analysis and systematic review, Goldstein et al13 demonstrated MIS TLIF had lower adverse events compared with open TLIF procedures (RR: 0.39; P=0.001). There are several limitations to this investigation. First, all patients were treated by a single surgeon at a single academic institution, which may not be generalizable to all orthopedic practices. Second, the data were retrospectively collected and inherently contains selection bias. Third, the diabetic cohort was not substratified by hemoglobin A1c levels and insulin-dependent status because of small sample size. Analyzing these factors may allow for a clearer association between the severity of diabetes and postoperative outcomes. Fourth, hospital readmission rates and revision surgeries were not tracked between cohorts. MIS TLIF requires a relatively short hospital stay and readmission or revision surgeries could have been useful for identifying complications after discharge.
Our study was limited in accounting for other confounders. Some possible confounders include diabetic protocols, postoperative care plans, and variable patient adherence with preoperative diabetic treatments. Although all cost data were collected at the same time, the fees for any service could have changed over 8 years. During this time, inflation, changes in hospital/payor relationships, or other financial influences may have led to cost fluctuations that were not captured in the current study. Future studies should aim to investigate cost variation over time whether because of inflation, changes in fees, or any other economic forces. Moving forward, a prospective investigation with larger sample size and further details regarding the diabetic status and medical management on a per-case basis are needed to better establish the risk for higher complications, inpatient length of stay, and hospital costs in patients undergoing MIS TLIF.
In this investigation, diabetes was not associated with higher postoperative complication rates, inpatient length of stay, or direct hospital costs after primary, single-level MIS TLIF. The reduced extent of operative exposure and tissue trauma in MIS TLIF may mitigate the risk of complications in diabetic patients, possibly preventing extensions in hospital stay length and associated hospital costs. Further investigation of the effect of diabetes on other perioperative and postoperative outcomes is required to fully evaluate it as a risk factor for minimally invasive lumbar surgery.
1. Golinvaux NS, Varthi AG, Bohl DD, et al. Complication rates following elective lumbar fusion in patients with diabetes: insulin dependence makes the difference. Spine (Phila Pa 1976). 2014;39:1809–1816.
2. Gruskay JA, Fu M, Bohl DD, et al. Factors affecting length of stay after elective posterior lumbar spine surgery: a multivariate analysis. Spine J. 2015;15:1188–1195.
3. Guzman JZ, Iatridis JC, Skovrlj B, et al. Outcomes and complications of diabetes mellitus
on patients undergoing degenerative lumbar spine surgery. Spine (Phila Pa 1976). 2014;39:1596–1604.
4. Walid MS, Newman BF, Yelverton JC, et al. Prevalence of previously unknown elevation of glycosylated hemoglobin in spine surgery patients and impact on length of stay and total cost. J Hosp Med. 2010;5:E10–E14.
5. Hey HW, Hee HT. Open and minimally invasive transforaminal lumbar interbody fusion
: comparison of intermediate results and complications. Asian Spine J. 2015;9:185–193.
6. Gruskay JA, Fu M, Basques BA, et al. Factors affecting length of stay and complications after elective anterior cervical discectomy and fusion: a study of 2164 patients from the American College of Surgeons National Surgical Quality Improvement Project Database (ACS NSQIP). Clin Spine Surg. 2016;29:E34–E42.
7. Jauregui JJ, Elmallah RK, Harwin SF, et al. Characteristics and complications of super-obese patients who underwent total knee arthroplasty. Orthopedics. 2016;39:e800–e805.
8. Uhl RL, Rosenbaum AJ, Dipreta JA, et al. Diabetes mellitus
: musculoskeletal manifestations and perioperative considerations for the orthopaedic surgeon. J Am Acad Orthop Surg. 2014;22:183–192.
9. Guzman JZ, Skovrlj B, Shin J, et al. The impact of diabetes mellitus
on patients undergoing degenerative cervical spine surgery. Spine (Phila Pa 1976). 2014;39:1656–1665.
10. Epstein NE. Predominantly negative impact of diabetes on spinal surgery: a review and recommendation for better preoperative screening. Surg Neurol Int. 2017;8:107.
11. Maloney PR, Halasz SR, Mallory GW, et al. The effect of diabetes mellitus
on 30-day outcomes following single-level open lumbar microdiscectomy: an aged-matched case-control study. J Neurosurg Sci. 2017;61:1–7.
12. Browne JA, Cook C, Pietrobon R, et al. Diabetes and early postoperative outcomes following lumbar fusion. Spine (Phila Pa 1976). 2007;32:2214–2219.
13. Goldstein CL, Macwan K, Sundararajan K, et al. Perioperative outcomes and adverse events of minimally invasive versus open posterior lumbar fusion: meta-analysis and systematic review. J Neurosurg Spine. 2016;24:416–427.
14. Worley N, Buza J, Jalai CM, et al. Diabetes as an independent predictor for extended length of hospital stay and increased adverse post-operative events in patients treated surgically for cervical spondylotic myelopathy. Int J Spine Surg. 2017;11:10.
15. Su AW, Habermann EB, Thomsen KM, et al. Risk factors for 30-day unplanned readmission and major perioperative complications after spine fusion surgery in adults: a review of the national surgical quality improvement program database. Spine (Phila Pa 1976). 2016;41:1523–1534.
16. Singh K, Nandyala SV, Marquez-Lara A, et al. A perioperative cost analysis comparing single-level minimally invasive and open transforaminal lumbar interbody fusion. Spine J. 2014;14:1694–1701.