As the elderly population increase, the incidence of degenerative spinal disease and its surgical treatment also increase, because development of anesthesiology and improvement of surgical instruments enabled aggressive surgical treatment in geriatric patients. However, there are controversies about clinical outcome, complication rates and mortality rates of geriatric patients after spinal surgery with the spinal problem.123456789101112131415161718 To our knowledge, there has been only one review article on the surgical outcome for geriatric patients in a geriatric journal,19 and no review article from the viewpoints of spine experts in the journal for spinal pathology.

Therefore, the purpose of the present study is to review the literature on the clinical outcome of surgical treatment in the geriatric patients with the diagnoses of lumbar stenosis or spondylolisthesis which are most commonly encountered in clinical practice of spinal surgeon.

A detailed computerized literature search using the MEDLINE search engine was employed using medical subject terms of "lumbar vertebrae", "surgery", and "geriatric". This database consists of literature published from January 1967 through December 2013. We included English language articles regarding clinical outcome, complication, or mortality after surgical treatment for geriatric patients older than 65 years old with spinal stenosis or spondylolisthesis. Articles regarding the surgical treatment for the geriatric patients with the diagnoses of disc herniation, scoliosis, fractures or infection were excluded. Review articles or case reports with less than ten patients were excluded. All full-text articles were read by two reviewers independently; each article was read by both the first author and one of the coauthors. In case of inconsistencies between two reviewers, the consensus was established through a discussion between them. We then categorized the articles as randomized clinical trial, non-randomized prospective study, and retrospective study for overall quality of the article. Finally, to focus on the most recent evidence, we limited our review to those published on or after 1986. We identified 39685 articles regarding lumbar surgery, among which 70 articles were about geriatric lumbar surgery and 21 dealt with geriatric lumbar surgery for spinal stenosis or spondylolisthesis. One article was excluded because it did not provide specific information about their diagnosis and age of study population,20 one other article was excluded because it dealt with patients younger than 65 years old,21 and another article was excluded because it included patients with heterogenous spinal diseases such as spinal stenosis, spondylolisthesis, herniated nucleus pulposus, instability, and scoliosis into the study population.22 Finally, we found one non-randomized prospective and seventeen retrospective studies that met the study criteria (Table 1,2,3,4). However, there was no randomized clinical trial that met the study criteria.

One non-randomized prospective and twelve retrospective studies dealt with clinical outcomes (Table 1).12345678910111213 The quality of the data appears to be poor except one non-randomized prospective study. The retrospective studies contained four case-control and eight case-series. Only eight studies used standard outcome measures in their study designs and this number is reduced to five if the visual analog scale (VAS) or neurologic function is not included.

Glassman, et al.1 investigated clinical outcomes of 224 patients following lumbar decompression and arthrodesis for spinal stenosis. Patients were divided into two groups according to ages; over (n=50) or below 65 years old (n=174).1 The health-related quality-of-life measures (HRQOL) including Oswestry Disability Index (ODI), the Medical Outcomes Study Short Form-36 (SF-36), back pain VAS, and leg pain VAS were evaluated for the mean follow-up of two years.1 There was no difference in the postoperative clinical outcomes.1 They performed another study on postoperative clinical outcomes of 178 patients following lumbar decompression and arthrodesis for spinal stenosis or spondylolisthesis.2 Patients were divided into two groups according to ages; over (n=85) or below 65 years old (n=93).2 There was no difference in the clinical outcomes with the mean follow-up of 33 months, evaluated with the same HRQOL.2

Okuda, et al.3 investigated clinical outcomes after lumbar decompression and fusion in 101 patients with spondylolisthesis. Patients were divided into two groups according to ages; over (n=31) or below 70 years old (n=70),3 and Japanese Orthopaedic Association score (JOA score) was evaluated at the mean follow-up of 50 months.3 The result indicated that the recovery rates in each group were 63% and 70%, respectively, showing no statistical difference.3

Arinzon, et al.4 performed the case-control study for the clinical outcomes after decompression in 235 patients with spinal stenosis. The authors included 83 patients over 75 years old in the study group and 152 patients between 65 and 75 years old in the control group.4 Clinical outcome was measured by patients' satisfaction at the mean follow-up of 42 months.4 Patients who were "not satisfied" were 10% in the study group and 15% in the control group, respectively, showing no difference.4

Rihn, et al.5 performed non-randomized prospective study on the clinical outcomes after decompression and fusion in patients with spinal stenosis or degenerative spondylolisthesis. The authors included 58 patients over 80 years old in the study group and 742 patients younger than 80 years old in the control group.5 The HRQOL outcomes including ODI, SF-36, stenosis bothersome index, low back pain bothersome scale, patient satisfaction, and patient self-reported improvement were measured at the final follow-up of 48 months.5 All HRQOL measures were similar between the study and control groups except the SF-36 bodily pain domain.5 The authors initially had the designs of both randomized cohort and observational cohort study. However, they found extensive crossover in the randomized cohort,5 so that they combined the randomized and observational cohorts into an "as-treated" analysis.5 In the present study, therefore, we classified it as the non-randomized prospective study.

Crawford, et al.6 performed the case series study for the clinical outcomes after decompression and fusion in patients over 75 years old with lumbar stenosis or spondylolisthesis. They included 35 patients over 75 years old in the study cohort.6 The HRQOL outcomes including ODI, SF-36, back pain VAS, and leg pain VAS were measured at the mean follow-up of 24 months.6 All HRQOL measures were found to improve until 2 years after operation.6

The other seven studies were the case series study, evaluated with non-standard outcome measures of VAS, satisfaction, independence and neurologic functions.78910111213 All of them showed that surgery favored the clinical outcome in geriatric patients with spinal stenosis or spondylolisthesis.78910111213

In summary, clinical outcome after spine surgery was satisfactory in geriatric patients with age between 65 and 101 after 2 to 8 years of follow-up period.12345678910111213 Addition of arthrodesis had no negative effect on the clinical outcomes.1236910

Complication was defined as any event for which the patient required a specific intervention. There was no randomized clinical trial regarding complications after spine surgery in geriatric patients. However, one non-randomized prospective study and ten retrospective studies on the complications were found, which can be considered as options (Table 2).345710111213141516 The quality of the data is poor except one non-randomized prospective study. The retrospective studies had three case-control and seven series studies.

One non-randomized prospective study and three case-control studies showed no statistical difference in complication rate after lumbar fusion in geriatric patients.34515 Four out of seven case series studies found that advanced age did not increase operation-associated complication,10111213 however, three out of seven case series studies found contrary results, showing the complication ranging from 52% to 70% (Table 2).71416

Postoperative complication was divided into minor and major complication (Table 3).14 Minor complications included urinary tract infection (UTI) and ileus which do not affect clinical outcome, whereas major complications included deep wound infection, pneumonia, and renal failure, which affect clinical outcome.14 Carreon, et al.14 found that 21% of geriatric patients had at least one major complication and 70% of them had at least one minor complication (Table 3).14 The major complication was found in 49% of the patients older than 75 years old.14 Shabat, et al.7 showed that 52% of the patients had complication.7 Most of them were minor (Table 3).7 However, Raffo and Lauerman16 found that major complications occurred in 7 out of 20 patients (35%).

In summary of the complications, there was no increasing tendency after spinal surgery in geriatric patients in one non-randomized prospective, three case-control and four case series studies.3451011121315 However, postoperative complications increased in three case series studies, especially in patients over 80 years old.71416 Fortunately, most postoperative complications after spine surgery in elderly patients were minor (Table 3).710111214

There was no randomized clinical trial or non-randomized prospective study in regard to mortality in spine surgery of elderly patients. The quality of the data is poor. There were two retrospective studies on the mortality (Table 4).1718 Oldridge, et al.17 compared postoperative mortality in 34418 elderly patients, who underwent lumbar surgery under the diagnosis of spinal stenosis or spondylolisthesis, with age-matched healthy group. Relative risk of expected mortality was lower than 1, which indicates that mortality was not increased by surgery.17 Kim, et al.18 investigated 10-year survival of elderly patients who underwent spinal surgery under the diagnosis of spinal stenosis and compared them with age- and gender-matched controls of general population. The standardized mortality ratios were 0.45 in patients aged 70 to 85 who underwent spinal surgery, showing lower mortality rate in the surgery group.18 To summarize the mortality data, more than 3500 elderly patients over 65 years old who underwent lumbar surgery were followed up for 1 to 10 years, and the result showed that mortality did not increase in elderly patients compared with age-adjusted control group who did not undergo spinal surgery.1718

Because of controversies on the clinical outcome, complication and mortality of spinal surgery in geriatric patients with the spinal stenosis or spondylolisthesis, we herein reviewed the literature about spinal surgery for geriatric patients, focusing on the diagnosis of spinal stenosis and spondylolisthesis which are most commonly encountered in a clinical practice of spinal surgeon.

Although it is generally known that spine surgery in elderly patients could deteriorate postoperative clinical outcomes, it does not affect surgical results.12345678910111213 It is not too far different from the clinical outcome of surgery for the deformity in the geriatric patients.23

There are some reports to suggest that postoperative complications are increased in elderly patients,7141620 whereas the others report that complications are not increased.3451011121315 Nevertheless, most of them were minor complications,710111214 and did not increase the mortality associated with the operation.1718 It is the same with the complications of surgery for the deformity in the geriatric patients.2324 The retrospective study of minimal 5-level spinal deformity surgery in 21 patients older than 75 years old showed that the overall complication was 62% and major complication was 38%.24 It is similar with the complications after surgery for the heterogenous spinal diseases in geriatric patients. Benz, et al.22 investigated the complication after decompression in 68 patients over 70 years old with heterogenous spinal diseases such as spinal stenosis, spondylolisthesis, herniated nucleus pulposus, instability, and scoliosis. The total complication rate was 40%, and major complications occurred only in 12% of patients.22

Generally, elderly patients have more comorbidities. As the number of preoperative comorbidities increase, more complications develop.1624 Mortality not related to surgical procedure increased, whereas mortality related to surgical procedure did not increase.1718 It is also same with the mortality after surgery for the deformity in geriatric patients.24 The correlation of preoperative morbidity with postoperative complication or mortality was investigated in elderly patients.1116 Raffo and Lauerman16 reported that patients over 80 years old develop postoperative complications 9.20 fold more because they had more preoperative morbidities. On the other hand, Ragab, et al.11 found that preoperative morbidity did not cause postoperative complications and advanced age did not increase the morbidity associated with the operation.

As in any study, the present investigation has several limitations. First, all the literatures reviewed are all retrospective except one non-randomized prospective study. There have been five randomized controlled studies comparing the clinical results of surgical and conservative treatment for spinal stenosis of patients with all ages (Table 5).2125262728 During 1 to 4 years of short-term follow-up, patients treated operatively had better clinical results with radicular and low back pain. During long-term follow-up such as 10 years, there were no significant differences in clinical results between surgery and conservative management for low back pain, because it is due to degenerative changes, although the patients treated operatively had better outcomes with radicular pain.2125262728 Second, there is no data on patients who were lost to follow-up. Third, the operative procedures of decompression or decompression & fusion were different among the studies mentioned. Fourth, surgery-related complications, such as hardware failure, fracture, progression of spinal deformity due to osteoporosis, and degenerative polyneuropathy, were not addressed. Fifth, our study could be improved by the methodology of meta-analysis. Unfortunately, we have only one prospective and five case-control studies, which might be insufficient to make a conclusion with meta-analysis. Additionally, outcome measurements were not consistent among available studies. Despite these shortcomings, to our knowledge, this is the first review on the surgical outcome of geriatric patients in the journal of spinal pathology.

In conclusion, there was no difference in clinical outcome, postoperative complication and mortality rates between geriatric and non-geriatric patients. However, there is insufficient evidence to make strong recommendations with regard to spinal surgery for geriatric patients with lumbar stenosis and spondylolisthesis. High-quality, prospective randomized studies on the spinal surgery for geriatric patients with lumbar stenosis and spondylolisthesis are needed.