Early postoperative complications after elective degenerative lumbar spine surgery in elderly patients

Overview

Objective:
Degenerative lumbar spine dis­ease is an increas­­ing problem nowadays, af­fect­­ing more and more people, especial­ly, the elderly population. The older age of patients is general­ly as­sociated with a higher rate of postoperative complications.

Patients and methods:
A retrospective analysis of patients aged 60 years or older who underwent the elective spine surgery at the Clinic of Neurosurgery at the University Hospital Martin from January 2015 to December 2016. Authors as­ses­sed the incidence of early postoperative complications after lumbar spine surgery in older patients and found a cor­relation between comorbidities, selected risk factors and the occur­rence of postoperative complications.

Results:
Over­al­l, 107 patients were as­ses­sed (48 men and 59 women). The incidence of postoperative complications was 24.3%. The incidence of major complications was 9.4%, the incidence of minor complications was 17.8%. 7.5% of patients needed revision surgery. 83.2% of patients underwent decompres­sive surgery alone while, 16.8% of patients underwent spinal surgery with fusion. The rate of complications was higher in those patients who underwent surgery with instrumentation in comparison to the decompres­sive surgery alone (23.6 vs. 27.8%; p = 0.7561). There was a statistical­ly significant as­sociation between the presence of type 2 diabetes mel­litus and the incidence of complications (p = 0.0082). The length of a surgical procedure strongly af­fected the occur­rence of postoperative complications (p = 0,001).

Conclusion:
Almost one quarter of patients aged 60 years or older developed postoperative complications with the predominance of minor complications. The study showed an increas­­ing rate of complications in patients who underwent a surgical procedure with instrumentation compared to patients with decompres­sive surgery only, but without statistical significance. We proved an as­sociation between the surgery length, the length of the hospital stay, the presence of type 2 diabetes mel­litus and the presence of postoperative complications.

Key words:
spine surgery – postoperative complications – degenerative spine – comorbidities

Introduction

It is well known that the number of elderly people is increasing, so there is also an as­sociated increase of age-related dis­eases, such as degenerative changes in lumbar spine [1,2]. An ag­­ing population requires that surgeons must increasingly consider operative intervention in elderly patients. At the time of surgery indication, it is important to as­sess the risk of complications incidence, especial­ly surgical ones, because when they occur, they usual­ly lead to reoperation, prolonged hospitalisation and drug use, economic consequences and a compromised postoperative outcome and benefits [3,4]. Precise knowledge on postoperative complications is very important and valuable for a surgeon, as well as for a patient. If we want to perform degenerative lumbar spine surgery safely, it also re­quires understand­­ing the factors that predict a succes­sful outcome as well as the complications. The occur­rence of postoperative complications in spine surgery varies widely in literature [5]. In this study, we as­sess the impact of risk factors (body mass index; BMI, the type of surgical procedure, the length of surgical procedure) and comorbidities on the occur­rence of postoperative complications. In the plan­n­­ing of spine surgery, there are dif­ferences in indications among the neurosurgeons and orthopaedic surgeons, usual­ly depend­­ing on the surgeon’s own experience. Surgery in older patients has its clear and proven postoperative benefits and a succes­sful outcome when it is appropriately indicated, but there is a question of overtreatment [6]. Specific indications for procedures in pain-related surgery are general­ly lacking, therefore, in procedure selection, the characteristics of a patient and a dis­ease may be outweighed by the individual preferences of a surgeon [7]. These choices are very important, especial­ly in older patients, because greater invasiveness is usual­ly as­sociated with more complications [8,9]. Lots of trials showed an equal ef­ficiency between decompres­sion only and decompres­sion with fusion. Preoperative risk assassment is crucial for successful outcomes after any spine surgery [10,11]. Many factors may contribute to the complications incidence [12,13]. Better information about postoperative complications and risk factors would help surgeons to estimate risks and benefits, and to make more individual decisions and indications [14– 16].

Patients and methods

In this study, we as­ses­sed all patients aged 60 years or older who had undergone the surgery due to the degenerative lumbar spine dis­ease in the period of 2 years (2015– 2016). We included patients with any kind of degenerative lumbar spine dis­ease –  degenerative disc dis­ease, lumbar spine stenosis, degenerative spondylolisthesis and lumbar spine segmental instability. The inclusion criteria were dia­gnosis and the patient’s age.

All patients received 2 g of cefazolin (600 mg of clindamycin, event.) 60 min before skin incision. Antibio­tics were administered until 24 h postoperatively. Prolonged antibio­tic microbial prophylaxis (48 h postoperatively) was administered, when the time of surgery exceeded 240 min.

Regard­­ing deepvein thrombosis pro­phy­laxis, graduated stockings, as well as low mo­lecular weight heparin were used in all patients.

Transpedicular screw insertion was con­trol­led by C-arm fluoroscopy (Phillips, Amster­-dam, The Netherlands) dur­­ing the surgery. Peroperative haemostasis was control­led by unipolar and bipolar coa­gulation, exces­sive bleed­­ing was control­led by topical haemostatic agents (Spongostan, Gelaspon). Suction drains were used in almost all patients (except five patients after one-level discectomy or hemilaminectomy). Drains were placed over the laminectomy/ hemilaminectomy as close as pos­sible to the thecal sac, ne­gative vacuum pres­sure was applied im­mediately after the wound closure. Drains were removed between the 1^st and 3^rd post­operative day, depend­­ing on the amount of suction.

Each patient with diabetes mellitus (DM) on insulin therapy or oral antidiabetics un­-der­went diabetologic preoperative exam­ination with preoperative, peroperative and postoperative treatment recommendations. Glucose concentration was recorded in a 6-h glycemic profile before surgery. The use of oral antidiabetics (especially biguanides) was discontinued before surgery, glyce­mia was controlled with short acting insulin.Only patients with an adequate compen­sation of DM (glucose levels within 4.0–8.0 mmol/l range) underwent surgical pro­cedure. Glycemic profile was examined postoperatively in all patients with DM, levelsof glycemia were controlled by standard insulin therapy, hyperglycemia was controlledby short acting insulin. Oral antidiabetics were restarted, when the patient has re­sumed adequate and regular oral intake.

We focused on the occur­rence of in-hospital postoperative complications us­­ing a retrospective analysis. There was no specific consensus in previous articles regard­­ing the definition of a postoperative com­plication in spine surgery. We used the most relevant scheme divid­­ing post­operative complications into major (adverse events with permanent sequelae, usual­ly requir­­ing a reoperation) and minor (adverse perioperative events with a transient ef­fect) [17]. We also as­ses­sed the mortality of patients and the neces­sity for further reoperation.

We conducted this study to determine an as­sociation between postoperative complications and patient’s comorbidities. We as­ses­sed all comorbidities from which patients, who underwent spine surgery, were suf­fering, and we separately evaluated the impact of comorbidities on the occur­rence of postoperative complications us­­ing a statistical analysis.

The antropometric factors of our interest were BMI and gender. We analysed if the time of surgery and the type of surgical procedure (the level of invasivenes­s) had influenced the occur­rence of postoperative complications. Accord­­ing to the level of invasivenes­s, we divided patients into two groups –  decompres­sion only and decompres­sion with fusion. Decompres­sion included any kind of discectomy, hemilaminectomy or laminectomy without fusion.

The results were as­ses­sed by descriptive statistics. Statistical significance of the results obtained was as­ses­sed by descriptive statistics, Student’s t-test and a Fisher’s test Data were assessed by GraphPad Software (GraphPad Software, La Jolla, CA, USA). Probability values of < 0.05 were considered to be statistical­ly significant.

Results

In the entire study, data from 107 patients aged 60 years or older were as­ses­sed (Fig. 1). The average age of included patients was 69.5 years. 48 patients (44.9%) were males, 59 patients (55.1%) were females. Among these patients, 72.9% had the dia­gnosis of spinal stenosis, 14.3% of patients had spondylolistesis, 65.4% had dia­gnosis of herniated or degenerative disc dis­ease, 14.1% had segmental instability.

Postoperative complications were divided into 2 main groups –  major and minor complications. Perioperative events lead­­ing to adverse sequelae, usual­ly requir­­ing a return to an operat­­ing room, were as­ses­sed as major complications. Other events, as­sociated with transient sequelae, were deemed as minor. In this study cohort, we reported 9 major complications in 8 patients (7.4%) and 25 minor complications in 19 patients (17.8%). Over­al­l, 26 patients (24.1%) had 34 complications (Tab. 1).

Out of the total cohort of 107 patients, 89 (83.2%) patients underwent only decompres­sive surgery. 21 (23.6%) patientsof this group had complications. 18 (16.8%)patients underwent surgery with instru­ment­ation, 5 developed postoperative com­plications (27.8%). Patients, in whom a sur­gical procedure was performed with instru­mentation were more likely to develop complications, but the difference was not statistical­ly significant (p = 0.7561).

We divided patients accord­­ing to the level of invasivenes­s. Patients with a higher number of levels of surgical procedure developed more complications, but the difference was not statistical­ly significant (Tab. 2).

Eight patients underwent revision surgery, 4 of them due to epidural hematoma, 1 due to cerebrospinal fluid leak, the next one due to the malposition of the transpedicular screw, another one due to the deep wound infection. The last patient underwent revision surgery due to postoperative mas­sive dis creherniation. Major complications were strongly as­sociated with reoperation. One patient with postoperative epidural hematoma was without the application of subfascialsuction drain.

We analysed an as­sociation between selected antropometric parameters (BMI, gender, age) and the occur­rence of com­plications. The mean age of patients with complications was 69.4 years, and without complications was 70.2 years (p = 0.5309). BMI was higher in patients with complications (29.2 vs. 28.9), but the difference was not statistical­ly significant (p = 0.8941). The presence of complications was higher in a cohort of female patients than in male patients (28.9 vs. 18.8%; p = 0.2630).

Comorbidities such as arterial hyper­tension, osteoporosis, DM and other systemic dis­eases were as­ses­sed to determine any cor­relation with postoperative complications. The comorbidities that mostly occur­red were evaluated separately. The most com­mon comorbidity was arterial hypertension (86% of patients). We found a statistical­ly significant as­sociation between the presence of type 2 DM and the occur­rence of postoperative complications (Tab. 3, 4). Ten patients with type 2 DM had postoperative complications, 12 patients were on oral antidiabetics, 8 patients were on insulin ther­apy. 33.3% of patients on oral antidiabetics and 75% of patients on insulin ther­apy had postoperative complications. Complications in patients with type 2 DM were wound dehiscence (4 patients), DM decompensation (4 patients), urinary infection (1 patient) and postoperative anaemia with blood transfusion (1 patient).

The average time of surgical procedure was 107.3 min. Patients with postoperative complications had longer surgery time (142.0 vs. 96.2 min), and this difference was statistical­ly significant (p < 0.001) (Fig. 2). We can consider the longer time of surgery as an important risk factor of postoperative complications. Postoperative complications were also divided accord­­ing to the time of surgery (Tab. 5).

The length of stay in hospital ranged from 7 to 51 days. The occur­rence of complications strongly af­fected the duration of hospitalisation. An average hospital stay in patients without complications was 9 days, whereas in patients with complications, the average hospital stay extended to 14 days (p < 0.001). On the basis of these results, we can also predict increased medical costs of the hospitalisation of patients with complications after spine surgery.

Discus­sion

Nowadays, postoperative complications of spine surgery in older patients represent an important object of study because of the improvement of health care services and grow­­ing medical costs [18,19]. Complications after surgery are also an important stress factor to patients themselves, patients’ families, as well as for surgeons [20]. This dis­ease usual­­ly af­fects the elderly part of the population, therefore, the surgeons who evaluate these patients must careful­ly consider the potential benefit and pos­sible risk factors, such as substantial comorbidities. When the surgical treatment of degenerative lumbar spine is succes­sful, it can produce a remarkable improvement in function, ambulation and activities of daily living.

The rate of postoperative complications in spine surgery widely dif­fers in literature. In our study, we focus­sed on major and minor complications. Our results confirmed that spine surgery in older patients car­ries relatively high risk of postoperative complications. This study revealed an over­all complication incidence of 24.1% , which is similar to previous studies [10– 12,20,21]. The over­all incidence of major complications was 7.4%, the incidence of minor com­plications was 17.8%. Cas­sinel­li et al revealed the incidence of 33.1%, with the predominance of minor complications [21], Button et al reported the complication incidence of 30% [22]. Findings by Deyo et al demonstrated an 18% complications rate in a similar group of patients [6]. The most com­mon major complication requir­­ing revision surgery was epidural hematoma. The most com­mon minor complications were DM decom­pensation and wound dehiscence.

In this study, the rate of postoperative complications was higher in patients who underwent surgery with instrumentation than in patients with decompres­sion only (27.8 vs. 23.6%), but without statistical significance. More complex procedures were as­sociated with the higher rate of complications, but in general, the evidence for better ef­ficacy of more complex procedures for degenerative lumbar spine dis­ease is lacking [23]. It is not surpris­­ing that more complex surgery is as­sociated with a higher complications risk because fusion usual­ly requires longer surgery time, more extensive dis­section and placement of implants [24]. A few studies proved equivalent ef­ficacy for decompres­sion alone vs. decompres­sion and fusion (with the absence of spondylolisthesis) [25]. These results revealed that it is very important for a surgeon to as­sess whether decompres­sion alone is suf­ficient enough, whether stabilis­­ing structures (facet joints, interspinous ligaments) should be preserved, when the fusion is indicated and to what extent instrumentation is needed [26].

This study showed a statistical­ly significant cor­relation between the time of surgery and the occur­rence of complications. This result is also as­sociated with the type of surgical procedure, because the time of surgical procedures with fusion was longer than the time of decompres­sive surgery only. There was a statistical­ly significant dif­ference between the length of hospital stay in patients with and without complications (9 vs. 14 days). These results can predict economic importance of the occur­rence of postoperative complications, because a longer hospital stay may significantly increase medical costs of hospitalisation.

There were no statistical­ly significant dif­ferences in gender distribution (female vs. male –  28.9 vs. 18.8%) and BMI (29.2 vs. 28.9) among patients who developed complications and those who did not. BMI was higher in patients with complications, but the difference was not statistically significant. Other trials showed various results, a few studies showed an as­sociation between obesity and complications, but other studies failed to prove this cor­relation [26– 28].

An as­sociation between comorbidities and complications in spine surgery is not conclusively demonstrated in literature. Other previous studies used various types of comorbidity indexes to evaluate an impact on the occur­rence of complications. In this study, we evaluated the impact of comorbidities separately and despite a dif­ferent type of evaluation, our results were similar to other previous trials [8,29– 34]. We divided comorbidities into 11 groups accord­­ing to their incidence and importance. This study revealed an as­sociation between the presence of the type 2 DM and the increased risk of complications. This result showed a statistical significance (p < 0.05) as wel­l. Management of glycemia before and dur­­ing surgery is very important, especial­ly in diabetic patients, because of negative ef­fects of surgical stress and anaesthesia on blood glucose levels. Nowadays, there is no evidence-based guideline indicat­­ing when to cancel surgery due to hyperglycaemia. Elective surgery should not be performed in patients with a compromised metabolic state. Careful glycaemic management in patients with preoperative diabetologic examination with given recom­mendations may re­duce morbidity and mortality and im­prove surgical outcomes. In our study, instead of adequate compensation of DM preoperatively and postoperatively, type 2 DM was a statistical­ly significant risk factor of postoperative complications [35,36].

Two of the patients developed spon­dylodiscitis after surgery, despite of standard preoperative and postoperative antibio­tic prophylaxis and aseptic rules. Both patients underwent surgical procedure with instrumentation. One of the patients was treated conservatively, the other one was treated surgical­ly. Our results were similar to other studies show­­ing the incidence of postoperative deep wound infection of 0.21– 3.6% [37]. A study by Hey et al reported a reduced risk of postoperative surgical site infection us­­ing prophylactic intraoperative local vancomycin powder in patients undergo­­ing instrumented spine surgery [38].

Incidence of postoperative epidural he­matoma is a serious complication. In our study, 4 patients had a symp­tomatic postoperative epidural hematoma lead­­ing to reoperation. Only 5 patients were without applications of subfascial suction drains. One patient from this group had post­operative epidural hematoma. Use of the suction drain facilitates the removal of an intrawound hematoma, however, several studies have reported that the suction drain does not prevent the development of such a complication. Study by Dong Ki Ahn et al showed, that suction drains functioned well before a coagulation nidus was formed. Drains should be placed as close as pos­sible to the thecal sac, a vacuum should be con­nected before the clott­­ing of extravascular blood. It is recom­mended not to use any materials, that activate platelet and facilitate coagulation of extravascular blood to prevent dysfunction of suction drains, but the type of prevention depends mainly on a surgeon‘s experience. Drains should be removed between the 1^st and the 3^rd day after surgery. Prolonged use of drains is considered as a risk factor of postoperative surgical site infections [39,40].

Complications incidence in spine surgery widely varies and may be influenced by many factors. Almost one quarter of elderly patients developed postoperative complications with the predominance of minor complications. On the basis of this study, we proved the cor­relation between the surgery length, the length of hospital stay, the presence of the type 2 DM and the incidence of postoperative complications. We equal­ly proved an increas­­ing rate of complications in those elderly patients who underwent surgical procedure with instrumentation than in the patients with decompres­sive surgery only. Spinal surgery in elderly patients car­ries a relatively high risk of postoperative complications, but more individual decisions at the time of surgery indication, the knowledge about risk factors lead­­ing to the higher complications rate and better information about patients may lead to serious improvement in healthcare services, better postoperative outcomes and a decrease in medical costs.

This work was supported by project The application of PACS (Picture Archiving and Communication System) in the research and development, ITMS 26210120004.

The authors declare they have no potential conflicts of interest concerning drugs, products, or services used in the study.

The Editorial Board declares that the manu­script met the ICMJE “uniform requirements” for biomedical papers.

Accepted for review: 9. 1. 2018

Accepted for print: 28. 5. 2018

doc. MUDr. Branislav Kolarovszki, PhD.

Clinic of Neurosurgery

Jessenius Faculty of Medicine in Martin

Comenius University in Bratislava

and Martin University Hospital

Kollárova 2

036 59 Martin

Slovakia

e-mail: kolarovszki@jfmed.uniba.sk