Elderly Patients Undergoing Minimally Invasive Transforaminal Lumbar Interbody Fusion May Have Similar Clinical Outcomes, Perioperative Complications, and Fusion Rates As Their Younger Counterparts

Abstract

Background 

Although several studies have suggested that minimally invasive transforaminal lumbar interbody fusion (MIS-TLIF) may be especially beneficial in the elderly population due to lower operative morbidity and faster postoperative recovery, there are limited studies investigating the functional outcomes, quality of life, and satisfaction in elderly patients after MIS-TLIF. Furthermore, existing studies had substantial clinical, diagnostic, and surgical heterogeneity.

Questions/purposes 

We asked if elderly patients could experience comparable (1) patient-reported pain, disability and quality of life, (2) perioperative complications, and (3) radiological fusion rates as their younger counterparts after MIS-TLIF.

Methods 

Prospectively collected registry data of patients undergoing primary, single-level, MIS-TLIF for degenerative spondylolisthesis between 2012 and 2014 were reviewed. We included 168 patients, 39 of whom were at least 70 years old. Of the 129 patients younger than 70 years old, propensity-score matching was used to select 39 younger controls with adjustment for sex, BMI, American Society of Anesthesiologists score, and baseline clinical outcomes. Perioperative complications and radiologic data were compared.

Results 

There was no difference in back pain (mean difference -0.3 [95% confidence interval -1.0 to 0.5]; p = 0.52); leg pain (mean difference -0.1 [95% CI to 0.6-0.5]; p = 0.85); Oswestry Disability Index (mean difference -2.9 [95% CI -8.0 to 2.2]; p = 0.26); and SF-36 physical (mean difference 3.0 [95% CI -0.7 to 6.8]; p = 0.107); and mental component summary (mean difference 1.9 [95% CI -4.5 to 8.2]; p = 0.56); up to 2 years postoperatively; 85% of younger patients and 85% of elderly patients were satisfied (p > 0.99) while 87% and 80%, respectively, had fulfilled expectations (p = 0.36). Four perioperative adverse events occurred in each group. There was also no difference in the rate of fusion (87% in younger patients and 90% in elderly patients; p = 0.135).

Conclusions 

When clinical and surgical heterogeneity were minimized, elderly patients undergoing minimally invasive transforaminal lumbar interbody fusion not only had comparable rates of perioperative complications but also experienced similar improvements in pain, function, and quality of life. A high rate of satisfaction was achieved.

Level of Evidence 

Level II, prognostic study.

Introduction

As the population ages, the number of elderly patients undergoing surgery for degenerative spinal conditions will continue to increase. Spine aging results in degenerative changes such as spinal stenosis and degenerative spondylolisthesis, which are among the leading causes of pain, functional impairment, and decline in quality of life in the geriatric population [⁴⁴]. Transforaminal lumbar interbody fusion (TLIF) is a prevalent surgical procedure to treat degenerative spondylolisthesis and spinal stenosis [^{50, 51}]. However, many elderly patients with degenerative spinal disorders are often not considered candidates for surgery because of medical comorbidities, ageism, or subjective impressions of a patient’s fitness for spinal surgery [⁴³]. Such patients are also often reluctant to undergo spinal surgery based on the belief that older patients are at a higher risk of having serious surgical complications or a poor outcome after surgery.

A past study demonstrated a nearly threefold increase in the risk of postoperative complications in patients 75 years or older compared with younger patients during open spinal surgery [¹¹]. However, recent advances in microscopy, tissue retractors, and other specialized instruments have enabled surgeons to perform minimally invasive spine (MIS) surgery with percutaneous incisions and reduced soft-tissue disruptions. MIS-TLIF has been shown to have comparable clinical outcomes and fusion rates to open TLIF [^{16, 34, 47}], along with the often-cited benefits of less intraoperative blood loss [³⁹], reduced postoperative pain [³⁶], shorter hospital stay [²⁹], and lower complication rates [^{17, 18}]. Consequently, researchers have proposed that the reduced exposure-related morbidity associated with MIS surgery may benefit elderly patients, in whom intraoperative blood loss, postoperative mobilization, and wound healing have more profound implications [^{21, 22, 43}].

Although studies have suggested that MIS-TLIF is associated with fewer perioperative complications in elderly patients, there are limited studies investigating the functional outcomes and quality of life of elderly patients after MIS-TLIF [¹]. Furthermore, existing studies had substantial clinical, diagnostic, and surgical heterogeneity; hence, the isolated impact of age on clinical outcome remains unclear [^{6, 13, 37, 49}].

We aimed to compare the (1) patient-reported pain, disability and quality of life, (2) perioperative complications and (3) radiological fusion rates between elderly patients and younger controls undergoing MIS-TLIF.

Patients and Methods

This study was approved by a centralized institutional review board (2018/2356). Prospectively collected data of patients who underwent primary single-level MIS-TLIF for degenerative spondylolisthesis between 2012 and 2014 were retrospectively reviewed. The indications for surgery were Grade 1 or 2 degenerative spondylolisthesis with nerve compression symptoms including radicular pain, paraesthesia, or neurogenic claudication. Patients were assessed preoperatively using erect and dynamic flexion and extension lumbar spine radiographs and MRI. In addition, all patients did not respond to a trial of nonsurgical treatment for 3 months before undergoing surgery.

All procedures were performed by the senior authors (CMG, WMY, SBT) using a previously described technique [³⁴]. First, the operative level was confirmed using a mobile C-arm x-ray. A surgical incision was then made 3 cm to 5 cm parallel to the midline on the symptomatic side. Tissue dilators were inserted down to the facet complex. Facetectomy was done to visualize the posterolateral part of the intervertebral disc, after which, discectomy was performed and the endplates were prepared. Intradiscal spreaders were used to distract the disc space and allograft bone was placed anterior and contralateral to the annulotomy together with an interbody cage filled with bone graft. Cage positioning was then confirmed using fluoroscopy. To ensure decompression, the remainder of the ipsilateral facet and lamina was resected and the lateral margin of the ligamentum flavum was removed to expose the ipsilateral exiting and transversing nerve roots. If there was bilateral disease, the patient was tilted and the tubular retractor angled medially to visualize the contralateral side, followed by over-the-top decompression where indicated. After decompression, a percutaneous pedicle screw and rod was inserted via the same incision and a second construct via a contralateral incision. Compression was applied, then the construct was tightened to restore lordosis. Hemostasis and wound irrigation were done before closure.

An independent healthcare professional (WY) performed the preoperative and postoperative patient assessments. Demographic data, medical comorbidities, and the American Society of Anesthesiologists score were collected. Patients were assessed using the VAS for back pain and leg pain, the North American Spine Society score for neurogenic symptoms, the Oswestry Disability Index, and the SF-36. The operative time and length of stay were also recorded. The medical outcome study approach proposed by Mchorney and Ware [²⁸] was used to derive two higher-order summary scores for the SF-36: physical component score and mental component score.

Preoperative characteristics were recorded for all patients (Table 1). Of the 168 included patients, 39 patients were at least 70 years old. Propensity-score matching was used to select a younger group of patients aged younger than 70 years, with adjustment for potential confounding variables such as sex, BMI, American Society of Anesthesiologists score, and baseline clinical scores (back pain, leg pain, neurogenic symptoms, Oswestry Disability Index, physical component score, and mental component score) (Table 2). This enabled us to control for selection bias and maintain covariate balance by matching the smaller elderly group (n = 39) with a subset of the younger group (n = 129). It has been well-described for use in observational studies [²].

Table 1.: Preoperative characteristics of the groups before propensity-score matching (n = 168)

Table 2.: Preoperative characteristics of the groups after propensity score matching (n = 78)

Clinical improvement was defined using the minimal clinically important difference (MCID), which represents a change in score compared with baseline that is considered meaningful improvement to the patient. Published MCID threshold values of 14.9 for the Oswestry Disability Index, 2.1 for back pain, 2.8 for leg pain [³³], and 4.9 for the physical component score [¹⁴] were used to determine if MCID was achieved. All outcome scores were evaluated again at 1 month, 3 months, 6 months, and 2 years postoperatively, together with an assessment of the patient’s fulfilment of expectations and overall satisfaction with the outcome of surgery using the North American Spine Society Lower Back Pain instrument [⁹]. We then categorized the scores into satisfied or unsatisfied and fulfilled or unfulfilled expectations (Table 3).

Table 3.: Evaluation of patient satisfaction and expectation fulfilment

We recorded perioperative adverse events, whether medical or surgical, using the validated Spine Adverse Events Severity System [⁴⁰]. Adverse events were classified into four grades: Grade 1 (no or minimal treatment or effect on length of stay), Grade 2 (treatment and/or an increased length of stay of 3 to 7 days but not associated with long-term sequelae), Grade 3 (treatment and/or increased length of stay of more than 7 days and are associated with long-term sequelae), and Grade 4 (resulting in mortality). In addition, any blood transfusion or 30-day readmission was recorded.

We analyzed fusion rates at 2 years, based on the grading system of Bridwell et al. [⁵]. Radiographic adjacent-segment degeneration was determined by comparing radiographs taken preoperatively and at most recent follow-up, irrespective of clinical symptoms. The following degenerative changes were identified: anterolisthesis or retrolisthesis of more than 3 mm, a decrease in the adjacent-segment disc height of more than 3 mm, or an intervertebral angle of flexion greater than 5° [^{31, 52}]. We also recorded signs of implant loosening or cage migration.

We performed a power analysis before this study. To detect an MCID of 14.9 points in the Oswestry Disability Index from a baseline score of 45 with an SD of 15, a sample size of at least 32 patients in each group would be required to achieve a power of 0.80. All continuous data are expressed as the mean and SD. Statistical analyses were performed using the SPSS software package, version 20.0 (SPSS Inc, Chicago, IL, USA). Baseline patient characteristics and clinical outcomes were compared between the groups using a chi-square test and t-test for categorical and continuous variables, respectively. We defined statistical significance at the 5% level (p ≤ 0.05).

Results

Patients in both groups showed an improvement in back pain, leg pain, neurogenic symptoms, Oswestry Disability Index, physical component score, and mental component score up to 2 years postoperatively (Fig. 1A-F). There was no difference in any patient-reported outcome measure at any timepoint between the groups (p > 0.05 for each) (Table 4). The rates of MCID attainment, satisfaction, and expectation fulfilment were also comparable (Table 5). At 2 years, 85% (33 of 39) of elderly patients and 85% (33 of 39) of younger patients were satisfied (p = 1.000), while 80% and 87%, respectively, had fulfilled expectations (p = 0.362).

Fig. 1 A-F: These graphs show patient-reported outcome measures: (A) back pain, (B) leg pain, (C) neurogenic symptoms, (D) Oswestry Disability Index, (E) physical component summary, and (F) mental component summary from preoperatively to 2 years postoperatively.

Table 4.: Comparison of clinical outcomes between groups at different time intervals (n = 78)

Table 5.: Attainment of the minimal clinically important difference, satisfaction, and expectation fulfilment at different time intervals (n = 78)

There were no differences in the total number of perioperative complications between younger and older patients. Four of 39 patients (10%) in each group experienced an adverse event (p = 1.000) (Table 6). In the younger group, one patient had a dural tear that was intraoperatively corrected without any complications postoperatively. Two patients had delirium because they overused opioid analgesia, and these patients were treated nonoperatively. One patient with popliteal deep vein thrombosis was treated with subcutaneous heparin. In the elderly group, one patient had a postoperative ileus that resolved with laxatives, atrial fibrillation developed in one patient on postoperative day 2 that spontaneously converted to a sinus rhythm after treatment with atenolol and amiodarone, one patient had urinary retention that resolved with a temporary Foley catheter, and one patient had soleal deep vein thrombosis that was also treated with subcutaneous heparin. No patients in this study underwent blood transfusion. In addition, no 30-day readmissions were recorded in either group. Two patients in the elderly group died at 3.8 and 4.3 years after surgery because of unrelated causes.

Table 6.: Perioperative complications in the two groups (n = 78)

In terms of radiographic outcomes, 87% (34 of 39) and 90% (35 of 39) of younger and older patients, respectively, had Grade 1 or 2 fusion. There was no difference in radiographic fusion rates between the two groups (p = 0.135). In addition, two of 39 patients in the elderly group and four of 39 patients in the younger group (p = 0.675) had evidence of radiologic adjacent-segment degeneration at a mean follow-up interval of 3.9 ± 1.5 years, for which one younger patient underwent L5-S1 microdecompression and discectomy at 4.4 years after the initial surgery. Two patients in each group had posterior cage migration seen on postoperative radiographs, and signs of screw loosening appeared in one patient in the younger group at 2.3 years. However, all five patients with implant-related complications were treated nonoperatively and did not undergo an additional surgical intervention.

Discussion

With an increasing life expectancy in populations globally, the number of elderly patients will continue to rise [¹²]. Moreover, technological improvements and the growing desire for a better quality of life and have led to an increase in the number of spinal procedures in this patient group. The goal of MIS surgery is to reduce the disruption of soft tissues and improve postoperative recovery [^{21, 22, 25, 34, 43}]. This is particularly beneficial in elderly patients, in whom operative morbidity and delayed postoperative mobility can increase surgical risks [^{21, 22}]. Although other studies have examined the impact of age on complication rates after spine surgery, few have evaluated patient-reported outcomes and satisfaction in an elderly population. Furthermore, heterogeneity in surgical procedures and diagnoses makes interpretation of existing data challenging. The main finding of this study was that patients aged 70 years and older undergoing single-level MIS-TLIF for degenerative spondylolisthesis had similar pain, function, quality of life and satisfaction as a propensity score-matched cohort of younger controls. The perioperative complication rate and radiological fusion rate were also similar between the groups.

This study has several limitations. This was a nonrandomized, comparative study. However, outcome measures were prospectively collected according to an established protocol and stored in a spine registry, likely addressing any bias in data collection. Furthermore, we attempted to reduce bias using propensity-score matching to adjust for possible confounding factors, such as sex, BMI, American Society of Anesthesiologists score and baseline clinical scores, thus ensuring a degree of homogeneity. Certain variables like the degree of canal or foraminal stenosis and extent of decompression could not be controlled and may have affected our results. The fact that MIS-TLIF was performed by a limited number of surgeons at our institution could have biased upward the results for both the young and elderly groups if all of these surgeons had excellent operative techniques. Further, although the study had sufficient statistical power, the small sample size may still limit the conclusions that can be drawn from this study. Because our prospectively maintained spine registry only records pre and postoperative parameters for patients who decide to undergo elective surgery for their spinal condition, some patients with the same diagnosis (that is, intractable pain, neurogenic symptoms, and functional limitations from degenerative spondylolisthesis) who opted for nonoperative treatment instead may not have been captured by our database, hence leading to potential selection bias. Consequently, we were also unable to assess the effect of age and/or other preoperative factors on the decision-making process when considering lumbar spine surgery, although this was not the primary aim of our study. Lastly, we did not delineate very elderly patients (at least 80 years old) from elderly patients as there were only four such patients in the cohort; therefore, it is possible that patients in their ninth decade may have poorer outcomes. Further studies on this subgroup of very elderly patients are warranted.

There is considerable variability in adverse events after surgical management of degenerative lumbar spinal disease in elderly people. Investigation of the perioperative outcomes of open lumbar spine surgery in relation to age have produced conflicting results [^{4, 6, 7, 10, 11, 23, 24, 37, 38, 45}]. Higher complication rates in elderly people have been reported in decompressive [^{4, 10, 23}] and fusion procedures [^{6, 11, 37}]. The complication rate was 41% complication rate in one study on patients older than 70 undergoing decompression or fusion [⁴], while another reported an 80% complication rate in patients older than 65 years who underwent posterior decompression and fusion [⁶], which increased with age, blood loss, length of operation, and number of levels. In contrast, other studies have shown similar results in elderly and nonelderly patients undergoing lumbar decompression and/or fusion [^{7, 24, 38, 45}]. Ragab et al. [³⁸] did not find increased morbidity in elderly patients who underwent surgery for spinal stenosis, with or without fusion. Furthermore, the addition of instrumentation did not increase the rate of perioperative complications [⁷]. Relatively fewer studies on MIS surgery with instrumentation in elderly patients have been conducted [^{21, 26, 41}]. A recent systematic review reported that very few studies investigating the efficacy of MIS approaches have specifically targeted an elderly population [⁴⁶]. Karikari et al. [²¹] studied 66 consecutive patients aged 70 years or older who underwent extreme lateral interbody fusion or MIS-TLIF and reported an overall low rate of major complications. Another study compared 40 patients who were at least 80 years old undergoing extreme lateral interbody fusion with 20 patients undergoing open posterior lumbar interbody fusion, and found a lower complication rate (8% versus 60%), blood loss transfusion rate (0% versus 70%), and hospital stay (1.3 versus 5.3 days) in the MIS group, which the authors attributed to the less invasive nature of the MIS procedure [⁴¹]. Lee and Fessler [²⁶] reported that 39 patients older than 65 years who underwent a single-level mini-open TLIF experienced good clinical and radiologic results with few perioperative complications at 3 years postoperatively; 20% of elderly patients (seven of 35) had a total of 12 complications (three major and nine minor). In our study, only 10% of elderly patients experienced a minor adverse event perioperatively, and we noted no major complications or blood transfusions. This finding of lower perioperative complication rates was consistent across all studies, highlighting the benefit of reduced exposure-related morbidity associated with MIS surgery in elderly patients.

Existing studies show that the assessment of success after spinal fusion surgery in the geriatric patient has been focused on perioperative complication rates, while little attention has been directed toward improvement in perceptions of pain, function, and quality of life or patient satisfaction after MIS surgery in elderly patients [^{1, 30, 41, 43}]. To our knowledge, only one comparative study has been conducted to date [¹]. Rosen et al. [⁴³] reported improvement in Oswestry Disability Index and SF-36 scores only in the first 3 months in their series of 57 patients with a mean age of 81 years undergoing MIS lumbar decompression for spinal stenosis, while another study only reported VAS pain scores, which decreased from a mean of 8.6 before surgery to 1.4 at 12 months [⁴¹]. However, no younger control group was present in these studies. More recently, a further study compared 41 elderly patients aged 70 years and older with younger patients who underwent minimally invasive decompression alone or decompression with instrumented fusion and reported no difference in 1-year Oswestry Disability Index scores or interval improvements in Oswestry Disability Index scores [¹]. However, their study was limited by the use of only a single outcome metric (Oswestry Disability Index) for comparison between the two groups, and another validated outcome measure such as the SF-36 or VAS may have yielded a different result. Follow-up was also limited to 1 year; hence, the authors acknowledged that a possible differential deterioration in Oswestry Disability Index scores between groups over a longer follow-up duration could occur. Although the two groups were homogeneous regarding clinical presentation, combining patients with spinal stenosis with and without degenerative spondylolisthesis introduced a degree of diagnostic heterogeneity. Moreover, among the 13 patients who underwent lumbar fusion, the number of levels fused (one or two levels) was also not controlled, leading to additional heterogeneity in surgical procedure. When clinical, surgical, and diagnostic heterogeneity were minimized, we found that older patients had comparable patient-reported outcome measure scores after MIS-TLIF. Propensity-score matching was used to adjust for baseline differences in the American Society of Anesthesiologists score and comorbidities, which was not done in a previous study [¹]. Using a longitudinal analysis for 2 years to determine the durability of postoperative changes in patient-reported outcome measure scores, we noted comparable results between the elderly and younger groups that were maintained over time with no deterioration. No difference was observed in physical component scores, a generic health-related quality of life metric. After controlling for baseline comorbidities between the groups, we found that older age did not affect a patient’s disease-specific score nor their generalized physical score. To comprehensively address questions about how suitable elderly patients are to undergo MIS-TLIF, the MCID was also studied. We deemed that patients who had scores that met the MCID perceived they had meaningful improvement in pain, function, and quality of life. The elderly group had comparable rates of MCID attainment for all patient-reported outcome measures, indicating that these patients may also experience substantial clinical benefit after MIS-TLIF.

In recent decades, attempts have been made to better evaluate the clinical outcomes of medical care from the patient's perspective [⁴²]. As part of these efforts, patient satisfaction has emerged as a metric for evaluating the quality of clinical care [¹⁵]. To date, few studies have studied the effect of age on subjective satisfaction after lumbar surgery [^{3, 19, 35, 38}]. A noncomparative study used 70 years old as a cut-off to study elderly patients who underwent lumbar surgery; the final satisfaction rate was as high as 92% [³⁸]. Another study noted that 90% of patients older than 70 years were satisfied 2 years after posterior fusion [³]. Phan et al. [³⁵] reported a similar patient satisfaction index in patients 64 years or older undergoing anterior lumbar interbody fusion. Similarly, another study found no difference in patient satisfaction scores between patients younger than and older than 65 years undergoing oblique lateral interbody fusion [¹⁹]. However, no studies that we know of have investigated the influence of age on subjective satisfaction after MIS-TLIF to date. Our study used the North American Spine Society scoring method, which is a standardized and validated outcome assessment instrument [⁹]. We found that age did not influence patient satisfaction after MIS-TLIF (85% rate of satisfaction in both groups). In addition to reduced pain, greater function, and improved quality of life after MIS-TLIF, elderly patients had their expectations fulfilled and were just as satisfied with their outcome as their younger counterparts.

In addition to the risk of screw loosening, the fusion rate after spinal instrumentation is an important consideration when deciding between surgical and nonsurgical management for these elderly patients. However, few studies have evaluated this aspect of surgery in elderly people and previous studies had conflicting results. Noncomparative studies noted stable bony fusion rates of 80% to 90% in elderly patients undergoing posterolateral fusion [^{8, 48}]. In a study comparing 31 patients older than and younger than 70 years, there was a higher rate of collapsed union and delayed union in the elderly group undergoing posterior lumbar interbody fusion, although clinical results were similar [³²]. Likewise, another study reported a higher incidence of nonunion and new osteoporotic fractures in 38 octogenarians undergoing posterolateral fusion [²⁷]. In contrast with these findings, Jo et al. [²⁰] could not find a difference in fusion rates between patients older (92%) and younger (95%) than 65 years undergoing open TLIF. Consistent with this, our study on MIS surgery found no difference in Bridwell fusion grades (87% versus 90%) between elderly patients and younger patients at 2 years postoperatively.

As the population ages, physicians will encounter more elderly patients who have the desire to maximize their mobility and independence. Traditional assumptions about their expectations should be revisited in light of recent technological advancements that have the potential to improve their quality of life while maintaining high safety standards. Elderly patients undergoing MIS surgery not only have comparable perioperative complication rates to those of younger patients but also experience a clinically meaningful improvement in pain, function, and quality of life after MIS-TLIF. These results were sustained for 2 years with no longer-term deterioration, and a high rate of satisfaction and expectation fulfilment was also achieved.