Eur J Orthop Surg Traumatol DOI 10.1007/s00590-014-1556-3
ORIGINAL ARTICLE
Lumbar spine surgery in patients 80 years of age or older: morbidity and mortality L. Balabaud • S. Pitel • I. Caux • C. Dova B. Richard • P. Antonietti • C. Mazel
•
Received: 11 August 2014 / Accepted: 13 October 2014 Ó Springer-Verlag France 2014
Abstract Introduction The life expectancy increased significantly during last four decades and is associated with an increasing quality of life. The purpose of this study was to determine morbidity and mortality of degenerative lumbar spine surgery in patients 80 years of age or older. Methods A consecutive retrospective review evaluated 121 consecutive patients, who had undergone posterior decompression associated or not with lumbar instrumentation. The average age at the time of surgery was 83.2 ± 2.4 years. The study analyzed the comorbidities, the operative procedure and postoperative morbidity and mortality. Results Blood loss increased significantly with instrumentation (538 vs. 280 mL, p \ 0.0001). The average operative time was 103 ± 38 min and increased significantly with instrumentation (131.2 vs. 84 min, p = 0.003), blood loss (p \ 0.0001) and dural tears (p = 0.05). Dural tears occurred in 19 patients (15.7 %) and were associated significantly with previous lumbar surgery (33.3 vs. 12.4 %, p = 0.017) and major complications (30 vs. 9.9 %, p = 0.01). Major complications occurred in 16 patients (13 %). Minor complications occurred in 36 patients (29.7 %). The delirium was associated with instrumentation (22 vs. 7 %, p = 0.017) and blood loss (520 vs. 348 mL, p = 0.034). The average hospital stay was 11.3 ± 8.1 days.
P. Antonietti: Deceased. L. Balabaud (&) S. Pitel I. Caux C. Dova B. Richard P. Antonietti C. Mazel Department of Orthopaedics, Institut Mutualiste Montsouris, 42 Boulevard Jourdan, 75014 Paris, France e-mail: [email protected] C. Mazel Paris 13 University-Sorbonne Paris Cite´, Paris, France
No patients died after a postoperative period of 1 year. The average follow-up was 24.3 ± 16.5 months. Conclusions The morbidity of degenerative lumbar spine surgery in patients 80 years of age or older was high. Blood loss, operative time, instrumentation, previous surgery and dural tears increased significantly the morbidity. This surgery must be decided very carefully and requires to inform the patient and his family of the high rate of complications. Keywords Complications Decompression Elderly Fusion Lumbar Morbidity
Introduction Life expectancy increased during the last four decades in western countries and is associated with an improvement of quality of life. Many octogenarians have an active life and want to preserve it. More and more, spine surgeons must properly evaluate the benefit risk ratio of spine surgery in octogenarians. Some studies [1–4] tried to evaluate morbidity and mortality, but they refer to short cohort or younger patients. To our knowledge, this is the first study evaluating morbidity and mortality of spine surgery in patients 80 years of age or older in a large cohort. The purpose of this study was to determine morbidity and mortality of lumbar spine surgery in patients 80 years of age or older and whether preoperative and perioperative factors correlated with this occurrence.
Materials and methods From January 2004 to August 2010, 121 consecutive patients 80 years of age or older who had undergone
123
Eur J Orthop Surg Traumatol
posterior decompression associated or not with a lumbar instrumentation were included in the study. This one was approved by the Institutional Review Board of our institute. We excluded all patients with fracture, neoplasm or spinal sepsis. The average age at the time of surgery was 83.2 ± 2.4 years (range 80–90 years). There were 48 men and 73 women. 24 patients (19.8 %) had undergone one or more previous surgical procedures on their lumbar spine. The preoperative diagnosis in the 121 patients is listed in the Table 1. Preoperative symptoms included radicular pain in 66 (54.5 %), neurogenic claudication in 50 (41 %) and cauda equine syndrome in 5 (4 %). Neurologic deficits were found in 44 patients (36 %), which included sensory deficit, motor weakness, or a combination of both. Five patients presented bowel and bladder dysfunction preoperatively. Fifty-four patients (44.7 %) took oral anticoagulants or platelet aggregation-inhibiting drugs, which were stopped or relayed by heparin preoperatively in usual period. The indications for surgery were lumbar spinal stenosis with or without spondylolisthesis and/or degenerative scoliosis, and failures of lumbar degenerative surgery, in patients with incapacitating back and radicular pain, no responsive to non-operative management. An instrumented fusion or dynamic stabilizations were always associated with the decompression when instability was assessed preoperatively. A laminectomy or foraminotomy was indicated for stenosis with preserved coronal and sagittal balance. For degenerative spondylolisthesis and/or preoperative assessed instability, an instrumented fusion (Fig. 1) or dynamic stabilizations were always performed. The use of Dynamic stabilization was performed for the limited degenerative spondylolisthesis. The rare long fusions were reserved to failures of previous lumbar surgery with coronal and/or sagittal imbalance. All the fusions were performed with pedicle screw instrumentation and autologous posterolateral cancellous iliac bone graft. Dynamic stabilizations (Dynesys, Zimmer, Warsaw, IN) were used in 13 patients (Fig. 2). The surgical procedures are listed in Table 2. Urinary tract catheters were put systematically in the patients with lumbar fusion in preoperative time. All procedures were performed by 1 of 3 senior orthopedic
Table 1 Preoperative diagnoses in the 121 patients Diagnoses
Numbers
Stenosis
51
Degenerative spondylolisthesis
45
Degenerative scoliosis
9
Complications of arthrodesis (pseudarthrodesis, junctional syndrome, vertebral body fracture below arthrodesis)
9
Disk hernia
7
123
spine surgeons (CM, PA, LB). Patient-controlled analgesia (PCA) with morphine pump for postoperative pain management was performed in 104 patients during 48 h, and morphine consumptions were recorded on postoperative days 1 and 2. Only three patients came back home, the other patients discharged to a nursing home or rehabilitation facility. A single observer (SP) who had no involvement with the patient’s care abstracted hospital records. The records of these 121 patients were analyzed for demographics, preoperative comorbidities (Charlson score, Table 3) [5] and risk assessment, primary diagnoses, and perioperative and postoperative complications. Among the preoperative clinical parameters, American Society of Anesthesiologists (ASA) classification of physical status was evaluated. Hospital records were studied looking for other hospitalizations or outpatient treatments for any complications, both medical and surgical, related to the index procedure within the first 6 months after discharge. According to Raffo and Carreon [2, 6], we considered as major complications: death, paralysis or neurologic injury, epidural hematoma, wound infection, pneumonia or pulmonary edema, a newonset cardiac arrhythmia, myocardial infarction, cerebrovascular accident, thromboembolic disease or gastrointestinal hemorrhage. Minor complications were defined as no life threatening and no compromise or dramatically change treatment, including delirium, ileus, urinary tract infection. According to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), diagnostic criteria for delirium are listed in Table 4. Blood transfusion was not considered as complication, but was recorded for descriptive analysis. For the analysis of the fusion, all the patients were followed up at 3, 6, 9 or 12 months (and each year) with systematic X-rays. If we have doubts as to the success of fusion, we did a follow-up CT scan. Statistical analyses Data were entered onto EXCEL software (Microsoft, Seattle, WA). Descriptive data are expressed as mean ± standard deviation. Data were analyzed using the SPSS 15.0 (Chicago, IL) software package. Chi-squared test, Fisher’s test, Student’s test and Pearson’s correlation coefficient test were used for the analysis. Blood loss had a normal distribution, and we used logarithmic transformation. Statistical significance was established at a P value \0.05.
Results Distribution of the 121 patients was as follows: one patient was ASA class I, 61 ASA class II, 57 ASA class III and 2
Eur J Orthop Surg Traumatol Fig. 1 Instrumented fusion (Twinflex, Alphatec Scient’x, France)
ASA class IV. One hundred and eight patients had preoperative comorbidities requiring medical intervention. The average Charlson score was 2.2 ± 1.6 (range 0–7). The average BMI was 25.6 ± 4.3 (range 17–38). The surgical procedures are listed in Table 2. An average of 2 levels (range 1–4 levels) was treated with decompression (71 procedures). An average of 2.2 levels (range 1–10 levels) was treated with decompression and instrumentation, fusion or dynamic stabilization (50 patients). The average estimated blood loss was 471 ± 394 mL (range 50–2,500 mL). Analysis of perioperative and postoperative transfusion requirements demonstrated that 36 patients (29.7 %) were treated by an average of 2.4 units of packed red blood cells (range 1–9). Blood loss increased significantly with or without instrumentation 668 ± 512 mL (range 150–2,500 mL) versus 333 ± 191 mL (range 50–900 mL) (p \ 0.0001). The average operative time was 103 ± 38 min (range 33–235 min). The average operative time increased significantly with instrumentation 131.2 ± 31.2 versus 84 ± 29 min (p = 0.003). The operative time was associated significantly with blood loss (p \ 0.0001) and dural tears (118 ± 47 vs. 100 ± 35 min, p = 0.05). Dural tears occurred in 19 patients (15.7 %). No major intraoperative complication was noted. Dural tears increased significantly with previous lumbar surgery (33.3 vs. 12.4 %, p = 0.017). Dural tears were associated significantly with major complications (30 vs. 9.9 %, p = 0.01). There was also a trend between major
complications and previous lumbar surgery (25 vs. 10.3 %, p = 0.067). Major complications occurred in 16 patients (13 %) and are listed in Table 5. One major complication occurred in 6 patients, two in 8 patients, three and four in 2 patients. Three of the four patients, who had epidural hematoma, presented persistent neurologic deficit. Among the wound infections (4 %), the identified bacteria were Escherichia coli (2 cases), Enterococcus faecalis (2 cases), and Staphylococcus aureus (1 case). Minor complications occurred in 36 patients (29.7 %) and are listed in Table 6. Delirium, urinary retention and urinary tract infection were found, respectively, in 16 patients (13 %), 10 patients (8 %) and 6 patients (5 %). The rate of urinary retention was probably underestimated because urinary tract catheters were systematically used in patients with lumbar fusion. The delirium was associated with instrumentation 22 versus 7 % (non-instrumented) (p = 0.017) and blood loss 520 mL (range 358–750 mL) versus 348 mL (range 304–398 mL, p = 0.034), respectively. PCA morphine consumptions on postoperative days 1 and 2 were 19.3 ± 14.2 mg (range 3–64 mg) and 11.7 ± 13.7 mg (range 0–55 mg), respectively. The daily PCA morphine consumptions were not responsible for confusion in any of these patients. The analysis of the fusion rate after a minimum of 1 year found 30 out of 37 patients. The average radiological follow-up was 2.05 ± 1.25 years (range 1–6 years). Among the 7 failures of fusion, instrumentations were performed in L4L5 (4 patients), L3L4, L3-
123
Eur J Orthop Surg Traumatol Fig. 2 Dynamic stabilizations (Dynesys, Zimmer, Warsaw, IN)
Table 2 Surgical procedures in the 121 patients Surgical procedures
Numbers
Non-instrumented surgeries
71
Laminectomies/foraminotomies
64
Discectomies ?Laminectomies
2 5
Instrumented surgeries Laminectomies/foraminotomies
50 32
?Posterolateral fusion and instrumentation
13
?Dynamic stabilization
3
Posterolateral fusion and instrumentation
1
Redux of posterolateral fusion and instrumentation ? osteotomy
1
123
sacrum and L1-sacrum. These failures of fusion were well tolerated. It was important to emphasize that only one patient out of the four with extended fusions (thoracolumbosacral fusion) developed a pseudarthrosis. Other failures resulted from adjacent level fractures or disk degeneration above the instrumentation. The average hospital stay was 11.3 ± 8.1 days (range 7–71 days), with a median of 6 days. The average follow-up was 24.3 ± 16.5 months (range 8–84 months). No patients died after a postoperative period of 1 year. No relation was found between the minor or major complications, and the gender, ASA classification, BMI, comorbidity Charlson score, anticoagulants, neurologic deficit, number of laminectomy levels and number of instrumented levels.
Eur J Orthop Surg Traumatol Table 3 Weighted index of comorbidity according to Charlson et al. [5]
Major complications
Numbers
Wound infection
5 (4 %)
Epidural hematoma
4 (3.3 %)
Congestive heart failure
Neurologic deficit
4 (3.3 %)
Peripheral vascular disease
New-onset cardiac arrhythmia
3 (2.5 %)
Cerebrovascular disease
Pneumonia and repiratory distress
3 (2.5 %)
Dementia
Congestive heart failure, coronaropathy
2 (1.6 %)
Chronic pulmonary disease
Thromboembolic disease
2 (1.6 %)
Connective tissue disease Ulcer disease
Renal failure
2 (1.6 %)
Pulmonary embolus
1 (0.8 %)
Mild liver disease
SIADH
1 (0.8 %)
Diabetes
Total
16 procedures (13 %)
Assigned weights for disease
Conditions
1
Myocardial infarct
2
Table 5 Major complications for the 121 patients
Hemiplegia Moderate or severe renal disease
SIADH syndrome of inappropriate antidiuretic hormone
Diabetes with end organ damage Any tumor Leukemia
Minor complications
Numbers
Moderate or severe liver disease
Confusion
16 (13 %)
Metastatic solid tumor
Urinary retention
10 (8 %)
AIDS
Urinary tract infection
6 (5 %)
Ileus
3 (2.5 %)
Hyponatremia
3 (2.5 %)
Hypokalemia
2 (1.6 %)
Digestive tract infection
1 (0.8 %)
Lymphoma 3 4
Table 6 Minor complications for the 121 patients
Assigned weights for each condition that a patient had. The total equals the score Table 4 Diagnostic criteria for delirium Disturbance of consciousness (i.e., reduced clarity of awareness of the environment) with reduced ability to focus, sustain, or shift attention A change in cognition (such as memory deficit, disorientation, language disturbance) or the development of a perceptual disturbance that is better accounted for by a preexisting, established, or evolving dementia The disturbance develops over a short period of time (usually hours to days) and tends to fluctuate during the course of the day Evidence is shown from the history, physical examination or laboratory findings that the disturbance is caused by the direct physiologic consequences of a general medical condition From Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Washington, DC: American Psychiatric Association, 1994
A comparative synthesis between instrumented and noninstrumented surgeries is presented in Table 7 and enhances the major role of instrumentation in the occurrence of complications.
Discussion The increasing life expectancy and the necessity to preserve a good quality of life for more and more
Morphine intoxication
1 (0.8 %)
Total
36 procedures (29.7 %)
Digestive tract infection: clostridium difficile
octogenarians lead to suggest surgical treatment to patients with degenerative lumbar spine pathologies. However, the surgical treatment of degenerative lumbar pathologies among older patients is debatable. It is clear that non-surgical treatment must be favored, but patients not responding favorably to it require surgical treatment [7]. Decompression is often enough, but fusion with instrumentation cannot be avoided in patients with associated spine instability [8, 9]. Moreover, lumbar spine surgery in the elderly patients allows good functional results [10–16]. Some studies [1–4] refer to very short cohort with an average age inferior to 70 years and sometimes less. They give no significant answers to our questions. No large studies being available, the purpose of our study is to evaluate morbidity and mortality of lumbar spine surgery in elderly patients. In orthopedic surgery, surgeons have no real difficulties to decide for a total hip arthroplasty in patients aged 80 years and over. Nevertheless, Reindl et al. [17] compared the complication rates between lumbar spine surgery with or without instrumentation and total hip arthroplasty
123
Eur J Orthop Surg Traumatol
et al. [18] found a 12.17 % complication rate and a 0.17 % mortality rate after lumbar laminectomy. But in patients older than 85 years, these rates increased between 14.7 and 18.9 % complication rate, and between 0.22 and 1.4 % mortality rate, without and with three associated comorbidities, respectively. The authors concluded that complication and mortality rate in patients undergoing lumbar laminectomy for degenerative stenosis increase significantly with age and number of comorbidities. In the same kind of study, Deyo et al. [19] identified in the State of Washington hospital discharge registry 18 % complication rate and 0.6 % mortality rate for patients older than 75 years. Comparatively, these rates were, respectively, 9 and 0.07 % for the overall patients in the study. Among the literature, other studies evaluating morbidity and mortality in elderly patients are listed in Table 8. These results are very different, and the significant factor that seems to make the difference is the added fusion with or without instrumentation. Some authors have found a link between instrumented fusion and complication rates. Raffo and Lauermann [2] reported a 35 % major complication rate for patients older than 80 years in 75 % instrumented surgery. Although the series was short with only 20 patients, the authors correlated comorbidity and major complication rate. Carreon et al. [6] reported a 21 % major complication rate with 10 % wound infection, and a 70 % minor complication rate with 34 % urinary tract infection, for patients undergoing decompression and instrumented fusion. The complication rate increased significantly with older age, blood loss, longer operative time and the number of levels of arthrodesis. On the contrary, Jo¨nsson and Stro¨mqvist [12] reported a 4 % major complication rate for patients older than 70 years with decompression alone. Ragab et al. [20] were the only authors who found no increased morbidity in patients 70 years of age or older
Table 7 Comparative synthesis between instrumented and noninstrumented surgeries Global study
Noninstrumented
Instrumented
Surgeries
121
71
50
Sex (M/F)
48/73
29/42
19/31
Age (years) Comorbidities (Charlson)
83.2 ± 2.4 2.2 ± 1.6
83.6 ± 2.6 2.2 ± 1.5
82.5 ± 1.8 2.3 ± 1.7
BMI
25.6 ± 4.3
25.2 ± 4.3
26.1 ± 4.4
Operative time (min)
103 ± 38
84 ± 29
131.2 ± 31.2 (p = 0.003)
Blood loss (mL)
471 ± 394
333 ± 191
668 ± 512 (p \ 0.0001)
Dural tears
15.7 %
19.8 %
12 %
Major complications
13 %
11 %
16 %
Minor complications
29.7 %
26.7 %
36 %
Mental confusion
13 %
7%
22 % (p = 0.017)
1-day PCA morphin
19.3 ± 14.2
19.7 ± 14.9
18.9 ± 13.3
2-day PCA morphin
11.7 ± 13.7
11.1 ± 13.1
12.6 ± 14.6
Hospital stay (days)
11.3 ± 8.1
10.6 ± 8.1
12.2 ± 8.2
The averages of blood loss are not modified by logarithmic distribution in this table. 1 and 2-days PCA morphin in mg
in the elderly and found no significant difference. Do patients over the age of 80 that undergo spine surgery also have similar findings? Our first step is to assess the complication rate and mortality of lumbar spine surgery. In a retrospective cohort study using the National Inpatient Sample Database, Li
Table 8 Studies of the literature evaluating morbidity and mortality in elderly patients References
Number of patients
Age (years)
Previous surgery (%)
Instrumented fusion (%)
Operative time (min)
Blood loss (mL)
Minor complications
Major complications (%)
Jo¨nsson and Stro¨mqvist [12]
50
[70
18
0
140
660
–
4
Carreon et al. [6]
98
72
0
100
294
679
21 %
70
74.1 ± 5.5 (72.4 ± 4.9)
0
45
346 (254)
995 (863)
30.7 % (31.9 %)
3
Cassinelli et al. [21]
166
Ragab et al. [20]
118
74
Becker et al. [10]
195
76 ± 4
24
2
–
650
8%
2
–
100
–
–
18.9 %
14.7
Benz et al. [23]
68
76.5 ± 4.4
35
20.6
224
1,040
28 %
12
Vitaz et al. [24]
65
78
31
76
–
–
27 %
10
Raffo et al. [2]
20
[80
35
75
–
–
55 %
35
Rosen et al. [13]
57
80.8
–
0
–
–
30 %
1
19.8
41.3
103 ± 38
471 ± 394
29.7 %
13
Balabaud et al.
123
121
83.2 ± 2.4
Eur J Orthop Surg Traumatol
with only 2 % of instrumented fusion. They assessed with the usual criteria that the minor and major complication rates were 8 and 2 %, respectively. Becker et al. [10] reported 14.7 % major complication rate and 18.9 % minor complication rate for patients undergoing fusion with pedicle screws and rod instrumentation with or without intervertebral cages. The average age of those patients was 76 ± 4 years. For the authors, age is not a contraindication for instrumented fusion. Cassinelli et al. [21] reported 3 % major complication rate and 30.7 and 31.9 % minor complication rates in patients undergoing decompression with or without fusion, respectively. The authors emphasized that decompression associated with fusion of 4 or more segments increased significantly the occurrence of a major complication. In patients undergoing fusion, of an average age of 72 years, Deyo et al. [22] reported a complication rate 1.9 times greater than in those who had surgery without fusion. The blood transfusion rate was 5.8 times greater. The authors suggested that the indications for fusion among older patients could be improved. Benz et al. [23] reported 12 % major complication rate and 28 % minor complication rate. The authors emphasized no significant relationship between comorbidities and postoperative complications, whereas arthrodesis with or without instrumentation and greater anesthetic time were associated significantly with complications. Comparatively, Vitaz et al. [24] reported 10 % major complication rate and 27 % minor complication rate for patients older than 75 years. These results are interesting to compare to Rosen et al. [13] who reported 1 % major complication rate and 30 % minor complication rate for patients undergoing minimally invasive lumbar spine decompression. In a comparative study between open posterior interbody fusion (PLIF) and minimally invasive extreme lateral interbody fusion (XLIF), Rodgers et al. [25] found that complication rate, blood loss and hospital stay were significantly lower in minimally invasive surgery. Perhaps, minimally invasive lumbar spine surgery may improve the morbidity in octogenerians [25], but this consideration must be proved in a study with a greater number of patients. Excluding Cassinelli [21] and Rosen [13], who found lesser major complication rates, probably related to the differences in the studied population, all the studies seemed to confirm our results with minor complications near 30 % and major complications between 10 and 15 %. In our study, minor and major complications occurred in 29.7 and 13 %, respectively. These results were slightly higher than those of total hip arthroplasty in our department, because de Thomasson et al. [26] reported a 27 % complication rate in 72 patients aged 80 years or older. In fact, among the studies reporting lumbar surgery in the elderly patients, the influence of two points is debatable: fusion with instrumentation and comorbidities. The first debatable point is
the association between fusion with instrumentation and postoperative complication rate. In the present study, blood loss, longer operative time and instrumentation seemed to influence significantly the morbidity of degenerative lumbar spine surgery in elderly patients. These results confirm other studies [2, 6, 22, 23]. These three factors are directly correlated. Fusion with instrumentation increased significantly operative time (131.2 vs. 84 min) and blood loss (668 vs. 333 mL). Moreover, the increased operative time was associated with significant blood loss. In contrary to Carreon et al. [6], no relation was found between the complications and the number of instrumented levels. Moreover, to our knowledge, our study is the first to establish a link between delirium with instrumentation (22 vs. 7 %), and blood loss (520 vs. 348 mL), respectively. We agree with Ragab et al. [20] who emphasized that ‘‘aggressive intraoperative blood transfusion are required to decrease morbidity in elderly patients undergoing lumbar spine surgery.’’ The second debatable point concerns the influence of preoperative comorbidities. Some authors found that the number of comorbidities increased complication rate [2, 4, 18, 19, 27], while other demonstrated no correlation [6, 20, 23]. It seems logical that the number of comorbidities a patient has would correlate with the incidence of complications, but there was no relationship identified in this study. Is the number of comorbidities a real factor to influence complications? This finding may be due to our anesthesiologists’ stringent assessment of the patients preoperatively that eliminated the patients with a large number of comorbidities. In our study, wound infection occurred in 5 patients (4 %). This rate is similar to those reported in the literature contained between 0 and 15 % [1, 6, 8, 10, 12, 16, 17, 20, 24, 25] in the elderly patients and usually known in general population [14]. To our knowledge, the present study is the first to emphasize the prevalence of Gram negative bacteria in wound infection: E. coli (2 cases) and E. faecalis (2 cases), compared to one case of S. aureus. Whereas S. aureus is generally the cause of wound infections in younger patients following lumbar spine surgery, this series demonstrated that wound infections in patients over the age 80 years were more commonly caused by urinary tract or fecal organisms. This point is probably to compare in our study the prevalence of urinary retentions and urinary tract infections. Thus, we modified our preventive antibioprophylaxis and are most careful in the preoperative perineal preparation. In our study, dural tears occurred in 19 patients (15.7 %) and increased with previous surgery—33.3 versus 12.4 %—In the literature, this rate is very heterogeneous and contained between 0.003 and 16 % [1, 6, 8, 10, 12, 14, 17, 20, 21]. Some authors [2, 8, 12, 20, 23, 24] included
123
Eur J Orthop Surg Traumatol
patients who had undergone previous lumbar surgery, and they found no relation between dural tears and previous surgery. It is quite conceivable that previous surgery and the consequent dural adhesions could increase the risk of dural tears. Moreover, we found a link between major complications and dural tears. We hypothesize that this finding was due to the inability to drain the surgical site which favored the formation of an epidural hematoma. All of the patients with an epidural hematoma had a dural tear. Consequently, the avoidance of a dural tear is critically important to decrease postoperative morbidity specifically the operative time, blood loss and the development of a postoperative hematoma. Finally, although Olridge et al. [27] found an increasing rate of mortality in patients 80 years of age or older, our findings show that the incidence of mortality was not increased in the first year postoperatively. The main weakness of our study is its retrospective design. However, this study is continuous and all the patients had follow-up with a satisfactory time. The second weakness of this study was no comparison group which limited too affirmative conclusions. In conclusion, the morbidity of degenerative lumbar spine surgery in patients 80 years of age or older was high. Blood loss, operative time, instrumentation, previous surgery and dural tears increased significantly the morbidity. The lumbar spine surgery in patients 80 years of age or older must be decided very carefully and requires to inform the patient and his family of the high rate of morbidity. Instrumented fusion must be limited and more cautiously decided in older patients than in younger patients. Acknowledgments No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Conflict of interest
None.
References 1. Best NM, Sasso RC (2007) Outpatient lumbar spine decompression in 233 patients 65 years of age or older. Spine 32:1135–1139 2. Raffo CS, Lauerman WC (2006) Predicting morbidity and mortality of lumbar spine arthrodesis in patients in their ninth decade. Spine 31:99–103 3. Sanderson PL, Wood PLR (1993) Surgery for lumbar spinal stenosis in old people. J Bone Joint Surg Br 75-B:393–397 4. Smith EB, Hanigan WC (1992) Surgical results and complications in elderly patients with benign lesions of the spinal canal. JAGS 40:867–870 5. Charlson ME, Pompei P, Ales KL et al (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Cron Dis 40:373–383 6. Carreon LY, Puno RM, Dimar JR et al (2003) Perioperative complications of posterior lumbar decompression and arthrodesis in older adults. J Bone Joint Surg Am 85-A:2089–2092
123
7. Tadokoro K, Miyamoto H, Sumi M et al (2005) The prognosis of conservative treatments for lumbar spinal stenosis. Analysis of patients over 70 years of age. Spine 30:2458–2463 8. Greenfield RT, Capen DA, Thomas JC et al (1998) Pedicle screw fixation for arthrodesis of the lumbosacral spine in the elderly. Spine 23:1470–1478 9. Kazunori Y, Takashi S, Yosihisa K et al (1996) Indication of fusion for lumbar spinal stenosis in elderly patients and its significance. Spine 21:242–248 10. Becker P, Bretschneider W, Tuschel A et al (2010) Life quality after instrumented lumbar fusion in the elderly. Spine 35:1478–1481 11. Galiano K, Obwegeser AA, Gabl MV et al (2005) Long-term outcome of laminectomy for spinal stenosis in octogenarians. Spine 30:332–335 12. Jo¨nsson B, Stro¨mqvist B (1994) Lumbar spine surgery in the elderly. Complications and surgical results. Spine 19:1431–1435 13. Rosen DS, O’Toole JE, Eichholz KM et al (2007) Minimally invasive lumbar spinal decompression in the elderly: outcomes of 50 patients aged 75 years and older. Neurosurg 60:503–510 14. Weinstein JN, Lurie JD, Tosteson TD et al (2007) Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 356:2257–2270 15. Zheng F, Sandhu HS, Cammisa FP et al (2001) Predictors of functional outcome in elderly patients undergoing posterior lumbar spine surgery. J Spinal Disord 14:518–521 16. Crawford CH, Smail J, Carreon LY et al (2011) Health-related quality of life after posterolateral lumbar arthrodesis in patients seventy-five years of age and older. Spine 36:1065–1068 17. Reindl R, Steffen T, Cohen L et al (2003) Elective lumbar spinal decompression in the elderly: Is it a high-risk operation? Can J Surg 46:43–46 18. Li G, Patil CG, Lad SP et al (2008) Effects of age and comorbidities on complication rates and adverse outcomes after lumbar laminectomy in elderly patients. Spine 33:1250–1255 19. Deyo RA, Cherkin DC, Loeser JD et al (1992) Morbidity and mortality in association with operations on the lumbar spine. J Bone Joint Surg Am 74-A:536–543 20. Ragab AA, Fye MA, Bohlman HH (2003) Surgery of the lumbar spine for spinal stenosis in 118 patients 70 years of age or older. Spine 28:348–353 21. Cassinelli EH, Eubanks J, Vogt M et al (2007) Risk factors for the development of perioperative complications in elderly patients undergoing lumbar decompression and arthrodesis for spinal stenosis. Spine 32:230–235 22. Deyo RA, Ciol MA, Cherkin DC et al (1993) Lumbar spinal fusion. A cohort study of complications, reoperations, and resource use in the Medicare population. Spine 18:1463–1470 23. Benz RJ, Ibrahim ZG, Afshar P et al (2001) Predicting complications in elderly patients undergoing lumbar decompression. Clin Orthop 384:116–121 24. Vitaz TW, Raque GH, Shields CB et al (1999) Surgical treatment of lumbar spinal stenosis in patients older than 75 years of age. J Neurosurg 91:181–185 25. Rodgers WB, Gerber EJ, Rodgers JA (2010) Lumbar fusion in octogenarians. The promise of minimally invasive surgery. Spine 35:S355–S360 26. De Thomasson E, Caux I, Guigand O, Terracher R, Mazel C (2009) Total hip arthroplasty for osteoarthritis in patients aged 80 years or older: influence of co-morbidities on final outcome. Orthop Traumatol Surg Res 95:249–253 27. Oldridge NB, Yuan Z, Stoll JE et al (1994) Lumbar spine surgery and mortality among medicare beneficiaries, 1986. Am J Public Health 84:1292–1298
ORIGINAL ARTICLE
Lumbar spine surgery in patients 80 years of age or older: morbidity and mortality L. Balabaud • S. Pitel • I. Caux • C. Dova B. Richard • P. Antonietti • C. Mazel
•
Received: 11 August 2014 / Accepted: 13 October 2014 Ó Springer-Verlag France 2014
Abstract Introduction The life expectancy increased significantly during last four decades and is associated with an increasing quality of life. The purpose of this study was to determine morbidity and mortality of degenerative lumbar spine surgery in patients 80 years of age or older. Methods A consecutive retrospective review evaluated 121 consecutive patients, who had undergone posterior decompression associated or not with lumbar instrumentation. The average age at the time of surgery was 83.2 ± 2.4 years. The study analyzed the comorbidities, the operative procedure and postoperative morbidity and mortality. Results Blood loss increased significantly with instrumentation (538 vs. 280 mL, p \ 0.0001). The average operative time was 103 ± 38 min and increased significantly with instrumentation (131.2 vs. 84 min, p = 0.003), blood loss (p \ 0.0001) and dural tears (p = 0.05). Dural tears occurred in 19 patients (15.7 %) and were associated significantly with previous lumbar surgery (33.3 vs. 12.4 %, p = 0.017) and major complications (30 vs. 9.9 %, p = 0.01). Major complications occurred in 16 patients (13 %). Minor complications occurred in 36 patients (29.7 %). The delirium was associated with instrumentation (22 vs. 7 %, p = 0.017) and blood loss (520 vs. 348 mL, p = 0.034). The average hospital stay was 11.3 ± 8.1 days.
P. Antonietti: Deceased. L. Balabaud (&) S. Pitel I. Caux C. Dova B. Richard P. Antonietti C. Mazel Department of Orthopaedics, Institut Mutualiste Montsouris, 42 Boulevard Jourdan, 75014 Paris, France e-mail: [email protected] C. Mazel Paris 13 University-Sorbonne Paris Cite´, Paris, France
No patients died after a postoperative period of 1 year. The average follow-up was 24.3 ± 16.5 months. Conclusions The morbidity of degenerative lumbar spine surgery in patients 80 years of age or older was high. Blood loss, operative time, instrumentation, previous surgery and dural tears increased significantly the morbidity. This surgery must be decided very carefully and requires to inform the patient and his family of the high rate of complications. Keywords Complications Decompression Elderly Fusion Lumbar Morbidity
Introduction Life expectancy increased during the last four decades in western countries and is associated with an improvement of quality of life. Many octogenarians have an active life and want to preserve it. More and more, spine surgeons must properly evaluate the benefit risk ratio of spine surgery in octogenarians. Some studies [1–4] tried to evaluate morbidity and mortality, but they refer to short cohort or younger patients. To our knowledge, this is the first study evaluating morbidity and mortality of spine surgery in patients 80 years of age or older in a large cohort. The purpose of this study was to determine morbidity and mortality of lumbar spine surgery in patients 80 years of age or older and whether preoperative and perioperative factors correlated with this occurrence.
Materials and methods From January 2004 to August 2010, 121 consecutive patients 80 years of age or older who had undergone
123
Eur J Orthop Surg Traumatol
posterior decompression associated or not with a lumbar instrumentation were included in the study. This one was approved by the Institutional Review Board of our institute. We excluded all patients with fracture, neoplasm or spinal sepsis. The average age at the time of surgery was 83.2 ± 2.4 years (range 80–90 years). There were 48 men and 73 women. 24 patients (19.8 %) had undergone one or more previous surgical procedures on their lumbar spine. The preoperative diagnosis in the 121 patients is listed in the Table 1. Preoperative symptoms included radicular pain in 66 (54.5 %), neurogenic claudication in 50 (41 %) and cauda equine syndrome in 5 (4 %). Neurologic deficits were found in 44 patients (36 %), which included sensory deficit, motor weakness, or a combination of both. Five patients presented bowel and bladder dysfunction preoperatively. Fifty-four patients (44.7 %) took oral anticoagulants or platelet aggregation-inhibiting drugs, which were stopped or relayed by heparin preoperatively in usual period. The indications for surgery were lumbar spinal stenosis with or without spondylolisthesis and/or degenerative scoliosis, and failures of lumbar degenerative surgery, in patients with incapacitating back and radicular pain, no responsive to non-operative management. An instrumented fusion or dynamic stabilizations were always associated with the decompression when instability was assessed preoperatively. A laminectomy or foraminotomy was indicated for stenosis with preserved coronal and sagittal balance. For degenerative spondylolisthesis and/or preoperative assessed instability, an instrumented fusion (Fig. 1) or dynamic stabilizations were always performed. The use of Dynamic stabilization was performed for the limited degenerative spondylolisthesis. The rare long fusions were reserved to failures of previous lumbar surgery with coronal and/or sagittal imbalance. All the fusions were performed with pedicle screw instrumentation and autologous posterolateral cancellous iliac bone graft. Dynamic stabilizations (Dynesys, Zimmer, Warsaw, IN) were used in 13 patients (Fig. 2). The surgical procedures are listed in Table 2. Urinary tract catheters were put systematically in the patients with lumbar fusion in preoperative time. All procedures were performed by 1 of 3 senior orthopedic
Table 1 Preoperative diagnoses in the 121 patients Diagnoses
Numbers
Stenosis
51
Degenerative spondylolisthesis
45
Degenerative scoliosis
9
Complications of arthrodesis (pseudarthrodesis, junctional syndrome, vertebral body fracture below arthrodesis)
9
Disk hernia
7
123
spine surgeons (CM, PA, LB). Patient-controlled analgesia (PCA) with morphine pump for postoperative pain management was performed in 104 patients during 48 h, and morphine consumptions were recorded on postoperative days 1 and 2. Only three patients came back home, the other patients discharged to a nursing home or rehabilitation facility. A single observer (SP) who had no involvement with the patient’s care abstracted hospital records. The records of these 121 patients were analyzed for demographics, preoperative comorbidities (Charlson score, Table 3) [5] and risk assessment, primary diagnoses, and perioperative and postoperative complications. Among the preoperative clinical parameters, American Society of Anesthesiologists (ASA) classification of physical status was evaluated. Hospital records were studied looking for other hospitalizations or outpatient treatments for any complications, both medical and surgical, related to the index procedure within the first 6 months after discharge. According to Raffo and Carreon [2, 6], we considered as major complications: death, paralysis or neurologic injury, epidural hematoma, wound infection, pneumonia or pulmonary edema, a newonset cardiac arrhythmia, myocardial infarction, cerebrovascular accident, thromboembolic disease or gastrointestinal hemorrhage. Minor complications were defined as no life threatening and no compromise or dramatically change treatment, including delirium, ileus, urinary tract infection. According to Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), diagnostic criteria for delirium are listed in Table 4. Blood transfusion was not considered as complication, but was recorded for descriptive analysis. For the analysis of the fusion, all the patients were followed up at 3, 6, 9 or 12 months (and each year) with systematic X-rays. If we have doubts as to the success of fusion, we did a follow-up CT scan. Statistical analyses Data were entered onto EXCEL software (Microsoft, Seattle, WA). Descriptive data are expressed as mean ± standard deviation. Data were analyzed using the SPSS 15.0 (Chicago, IL) software package. Chi-squared test, Fisher’s test, Student’s test and Pearson’s correlation coefficient test were used for the analysis. Blood loss had a normal distribution, and we used logarithmic transformation. Statistical significance was established at a P value \0.05.
Results Distribution of the 121 patients was as follows: one patient was ASA class I, 61 ASA class II, 57 ASA class III and 2
Eur J Orthop Surg Traumatol Fig. 1 Instrumented fusion (Twinflex, Alphatec Scient’x, France)
ASA class IV. One hundred and eight patients had preoperative comorbidities requiring medical intervention. The average Charlson score was 2.2 ± 1.6 (range 0–7). The average BMI was 25.6 ± 4.3 (range 17–38). The surgical procedures are listed in Table 2. An average of 2 levels (range 1–4 levels) was treated with decompression (71 procedures). An average of 2.2 levels (range 1–10 levels) was treated with decompression and instrumentation, fusion or dynamic stabilization (50 patients). The average estimated blood loss was 471 ± 394 mL (range 50–2,500 mL). Analysis of perioperative and postoperative transfusion requirements demonstrated that 36 patients (29.7 %) were treated by an average of 2.4 units of packed red blood cells (range 1–9). Blood loss increased significantly with or without instrumentation 668 ± 512 mL (range 150–2,500 mL) versus 333 ± 191 mL (range 50–900 mL) (p \ 0.0001). The average operative time was 103 ± 38 min (range 33–235 min). The average operative time increased significantly with instrumentation 131.2 ± 31.2 versus 84 ± 29 min (p = 0.003). The operative time was associated significantly with blood loss (p \ 0.0001) and dural tears (118 ± 47 vs. 100 ± 35 min, p = 0.05). Dural tears occurred in 19 patients (15.7 %). No major intraoperative complication was noted. Dural tears increased significantly with previous lumbar surgery (33.3 vs. 12.4 %, p = 0.017). Dural tears were associated significantly with major complications (30 vs. 9.9 %, p = 0.01). There was also a trend between major
complications and previous lumbar surgery (25 vs. 10.3 %, p = 0.067). Major complications occurred in 16 patients (13 %) and are listed in Table 5. One major complication occurred in 6 patients, two in 8 patients, three and four in 2 patients. Three of the four patients, who had epidural hematoma, presented persistent neurologic deficit. Among the wound infections (4 %), the identified bacteria were Escherichia coli (2 cases), Enterococcus faecalis (2 cases), and Staphylococcus aureus (1 case). Minor complications occurred in 36 patients (29.7 %) and are listed in Table 6. Delirium, urinary retention and urinary tract infection were found, respectively, in 16 patients (13 %), 10 patients (8 %) and 6 patients (5 %). The rate of urinary retention was probably underestimated because urinary tract catheters were systematically used in patients with lumbar fusion. The delirium was associated with instrumentation 22 versus 7 % (non-instrumented) (p = 0.017) and blood loss 520 mL (range 358–750 mL) versus 348 mL (range 304–398 mL, p = 0.034), respectively. PCA morphine consumptions on postoperative days 1 and 2 were 19.3 ± 14.2 mg (range 3–64 mg) and 11.7 ± 13.7 mg (range 0–55 mg), respectively. The daily PCA morphine consumptions were not responsible for confusion in any of these patients. The analysis of the fusion rate after a minimum of 1 year found 30 out of 37 patients. The average radiological follow-up was 2.05 ± 1.25 years (range 1–6 years). Among the 7 failures of fusion, instrumentations were performed in L4L5 (4 patients), L3L4, L3-
123
Eur J Orthop Surg Traumatol Fig. 2 Dynamic stabilizations (Dynesys, Zimmer, Warsaw, IN)
Table 2 Surgical procedures in the 121 patients Surgical procedures
Numbers
Non-instrumented surgeries
71
Laminectomies/foraminotomies
64
Discectomies ?Laminectomies
2 5
Instrumented surgeries Laminectomies/foraminotomies
50 32
?Posterolateral fusion and instrumentation
13
?Dynamic stabilization
3
Posterolateral fusion and instrumentation
1
Redux of posterolateral fusion and instrumentation ? osteotomy
1
123
sacrum and L1-sacrum. These failures of fusion were well tolerated. It was important to emphasize that only one patient out of the four with extended fusions (thoracolumbosacral fusion) developed a pseudarthrosis. Other failures resulted from adjacent level fractures or disk degeneration above the instrumentation. The average hospital stay was 11.3 ± 8.1 days (range 7–71 days), with a median of 6 days. The average follow-up was 24.3 ± 16.5 months (range 8–84 months). No patients died after a postoperative period of 1 year. No relation was found between the minor or major complications, and the gender, ASA classification, BMI, comorbidity Charlson score, anticoagulants, neurologic deficit, number of laminectomy levels and number of instrumented levels.
Eur J Orthop Surg Traumatol Table 3 Weighted index of comorbidity according to Charlson et al. [5]
Major complications
Numbers
Wound infection
5 (4 %)
Epidural hematoma
4 (3.3 %)
Congestive heart failure
Neurologic deficit
4 (3.3 %)
Peripheral vascular disease
New-onset cardiac arrhythmia
3 (2.5 %)
Cerebrovascular disease
Pneumonia and repiratory distress
3 (2.5 %)
Dementia
Congestive heart failure, coronaropathy
2 (1.6 %)
Chronic pulmonary disease
Thromboembolic disease
2 (1.6 %)
Connective tissue disease Ulcer disease
Renal failure
2 (1.6 %)
Pulmonary embolus
1 (0.8 %)
Mild liver disease
SIADH
1 (0.8 %)
Diabetes
Total
16 procedures (13 %)
Assigned weights for disease
Conditions
1
Myocardial infarct
2
Table 5 Major complications for the 121 patients
Hemiplegia Moderate or severe renal disease
SIADH syndrome of inappropriate antidiuretic hormone
Diabetes with end organ damage Any tumor Leukemia
Minor complications
Numbers
Moderate or severe liver disease
Confusion
16 (13 %)
Metastatic solid tumor
Urinary retention
10 (8 %)
AIDS
Urinary tract infection
6 (5 %)
Ileus
3 (2.5 %)
Hyponatremia
3 (2.5 %)
Hypokalemia
2 (1.6 %)
Digestive tract infection
1 (0.8 %)
Lymphoma 3 4
Table 6 Minor complications for the 121 patients
Assigned weights for each condition that a patient had. The total equals the score Table 4 Diagnostic criteria for delirium Disturbance of consciousness (i.e., reduced clarity of awareness of the environment) with reduced ability to focus, sustain, or shift attention A change in cognition (such as memory deficit, disorientation, language disturbance) or the development of a perceptual disturbance that is better accounted for by a preexisting, established, or evolving dementia The disturbance develops over a short period of time (usually hours to days) and tends to fluctuate during the course of the day Evidence is shown from the history, physical examination or laboratory findings that the disturbance is caused by the direct physiologic consequences of a general medical condition From Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Washington, DC: American Psychiatric Association, 1994
A comparative synthesis between instrumented and noninstrumented surgeries is presented in Table 7 and enhances the major role of instrumentation in the occurrence of complications.
Discussion The increasing life expectancy and the necessity to preserve a good quality of life for more and more
Morphine intoxication
1 (0.8 %)
Total
36 procedures (29.7 %)
Digestive tract infection: clostridium difficile
octogenarians lead to suggest surgical treatment to patients with degenerative lumbar spine pathologies. However, the surgical treatment of degenerative lumbar pathologies among older patients is debatable. It is clear that non-surgical treatment must be favored, but patients not responding favorably to it require surgical treatment [7]. Decompression is often enough, but fusion with instrumentation cannot be avoided in patients with associated spine instability [8, 9]. Moreover, lumbar spine surgery in the elderly patients allows good functional results [10–16]. Some studies [1–4] refer to very short cohort with an average age inferior to 70 years and sometimes less. They give no significant answers to our questions. No large studies being available, the purpose of our study is to evaluate morbidity and mortality of lumbar spine surgery in elderly patients. In orthopedic surgery, surgeons have no real difficulties to decide for a total hip arthroplasty in patients aged 80 years and over. Nevertheless, Reindl et al. [17] compared the complication rates between lumbar spine surgery with or without instrumentation and total hip arthroplasty
123
Eur J Orthop Surg Traumatol
et al. [18] found a 12.17 % complication rate and a 0.17 % mortality rate after lumbar laminectomy. But in patients older than 85 years, these rates increased between 14.7 and 18.9 % complication rate, and between 0.22 and 1.4 % mortality rate, without and with three associated comorbidities, respectively. The authors concluded that complication and mortality rate in patients undergoing lumbar laminectomy for degenerative stenosis increase significantly with age and number of comorbidities. In the same kind of study, Deyo et al. [19] identified in the State of Washington hospital discharge registry 18 % complication rate and 0.6 % mortality rate for patients older than 75 years. Comparatively, these rates were, respectively, 9 and 0.07 % for the overall patients in the study. Among the literature, other studies evaluating morbidity and mortality in elderly patients are listed in Table 8. These results are very different, and the significant factor that seems to make the difference is the added fusion with or without instrumentation. Some authors have found a link between instrumented fusion and complication rates. Raffo and Lauermann [2] reported a 35 % major complication rate for patients older than 80 years in 75 % instrumented surgery. Although the series was short with only 20 patients, the authors correlated comorbidity and major complication rate. Carreon et al. [6] reported a 21 % major complication rate with 10 % wound infection, and a 70 % minor complication rate with 34 % urinary tract infection, for patients undergoing decompression and instrumented fusion. The complication rate increased significantly with older age, blood loss, longer operative time and the number of levels of arthrodesis. On the contrary, Jo¨nsson and Stro¨mqvist [12] reported a 4 % major complication rate for patients older than 70 years with decompression alone. Ragab et al. [20] were the only authors who found no increased morbidity in patients 70 years of age or older
Table 7 Comparative synthesis between instrumented and noninstrumented surgeries Global study
Noninstrumented
Instrumented
Surgeries
121
71
50
Sex (M/F)
48/73
29/42
19/31
Age (years) Comorbidities (Charlson)
83.2 ± 2.4 2.2 ± 1.6
83.6 ± 2.6 2.2 ± 1.5
82.5 ± 1.8 2.3 ± 1.7
BMI
25.6 ± 4.3
25.2 ± 4.3
26.1 ± 4.4
Operative time (min)
103 ± 38
84 ± 29
131.2 ± 31.2 (p = 0.003)
Blood loss (mL)
471 ± 394
333 ± 191
668 ± 512 (p \ 0.0001)
Dural tears
15.7 %
19.8 %
12 %
Major complications
13 %
11 %
16 %
Minor complications
29.7 %
26.7 %
36 %
Mental confusion
13 %
7%
22 % (p = 0.017)
1-day PCA morphin
19.3 ± 14.2
19.7 ± 14.9
18.9 ± 13.3
2-day PCA morphin
11.7 ± 13.7
11.1 ± 13.1
12.6 ± 14.6
Hospital stay (days)
11.3 ± 8.1
10.6 ± 8.1
12.2 ± 8.2
The averages of blood loss are not modified by logarithmic distribution in this table. 1 and 2-days PCA morphin in mg
in the elderly and found no significant difference. Do patients over the age of 80 that undergo spine surgery also have similar findings? Our first step is to assess the complication rate and mortality of lumbar spine surgery. In a retrospective cohort study using the National Inpatient Sample Database, Li
Table 8 Studies of the literature evaluating morbidity and mortality in elderly patients References
Number of patients
Age (years)
Previous surgery (%)
Instrumented fusion (%)
Operative time (min)
Blood loss (mL)
Minor complications
Major complications (%)
Jo¨nsson and Stro¨mqvist [12]
50
[70
18
0
140
660
–
4
Carreon et al. [6]
98
72
0
100
294
679
21 %
70
74.1 ± 5.5 (72.4 ± 4.9)
0
45
346 (254)
995 (863)
30.7 % (31.9 %)
3
Cassinelli et al. [21]
166
Ragab et al. [20]
118
74
Becker et al. [10]
195
76 ± 4
24
2
–
650
8%
2
–
100
–
–
18.9 %
14.7
Benz et al. [23]
68
76.5 ± 4.4
35
20.6
224
1,040
28 %
12
Vitaz et al. [24]
65
78
31
76
–
–
27 %
10
Raffo et al. [2]
20
[80
35
75
–
–
55 %
35
Rosen et al. [13]
57
80.8
–
0
–
–
30 %
1
19.8
41.3
103 ± 38
471 ± 394
29.7 %
13
Balabaud et al.
123
121
83.2 ± 2.4
Eur J Orthop Surg Traumatol
with only 2 % of instrumented fusion. They assessed with the usual criteria that the minor and major complication rates were 8 and 2 %, respectively. Becker et al. [10] reported 14.7 % major complication rate and 18.9 % minor complication rate for patients undergoing fusion with pedicle screws and rod instrumentation with or without intervertebral cages. The average age of those patients was 76 ± 4 years. For the authors, age is not a contraindication for instrumented fusion. Cassinelli et al. [21] reported 3 % major complication rate and 30.7 and 31.9 % minor complication rates in patients undergoing decompression with or without fusion, respectively. The authors emphasized that decompression associated with fusion of 4 or more segments increased significantly the occurrence of a major complication. In patients undergoing fusion, of an average age of 72 years, Deyo et al. [22] reported a complication rate 1.9 times greater than in those who had surgery without fusion. The blood transfusion rate was 5.8 times greater. The authors suggested that the indications for fusion among older patients could be improved. Benz et al. [23] reported 12 % major complication rate and 28 % minor complication rate. The authors emphasized no significant relationship between comorbidities and postoperative complications, whereas arthrodesis with or without instrumentation and greater anesthetic time were associated significantly with complications. Comparatively, Vitaz et al. [24] reported 10 % major complication rate and 27 % minor complication rate for patients older than 75 years. These results are interesting to compare to Rosen et al. [13] who reported 1 % major complication rate and 30 % minor complication rate for patients undergoing minimally invasive lumbar spine decompression. In a comparative study between open posterior interbody fusion (PLIF) and minimally invasive extreme lateral interbody fusion (XLIF), Rodgers et al. [25] found that complication rate, blood loss and hospital stay were significantly lower in minimally invasive surgery. Perhaps, minimally invasive lumbar spine surgery may improve the morbidity in octogenerians [25], but this consideration must be proved in a study with a greater number of patients. Excluding Cassinelli [21] and Rosen [13], who found lesser major complication rates, probably related to the differences in the studied population, all the studies seemed to confirm our results with minor complications near 30 % and major complications between 10 and 15 %. In our study, minor and major complications occurred in 29.7 and 13 %, respectively. These results were slightly higher than those of total hip arthroplasty in our department, because de Thomasson et al. [26] reported a 27 % complication rate in 72 patients aged 80 years or older. In fact, among the studies reporting lumbar surgery in the elderly patients, the influence of two points is debatable: fusion with instrumentation and comorbidities. The first debatable point is
the association between fusion with instrumentation and postoperative complication rate. In the present study, blood loss, longer operative time and instrumentation seemed to influence significantly the morbidity of degenerative lumbar spine surgery in elderly patients. These results confirm other studies [2, 6, 22, 23]. These three factors are directly correlated. Fusion with instrumentation increased significantly operative time (131.2 vs. 84 min) and blood loss (668 vs. 333 mL). Moreover, the increased operative time was associated with significant blood loss. In contrary to Carreon et al. [6], no relation was found between the complications and the number of instrumented levels. Moreover, to our knowledge, our study is the first to establish a link between delirium with instrumentation (22 vs. 7 %), and blood loss (520 vs. 348 mL), respectively. We agree with Ragab et al. [20] who emphasized that ‘‘aggressive intraoperative blood transfusion are required to decrease morbidity in elderly patients undergoing lumbar spine surgery.’’ The second debatable point concerns the influence of preoperative comorbidities. Some authors found that the number of comorbidities increased complication rate [2, 4, 18, 19, 27], while other demonstrated no correlation [6, 20, 23]. It seems logical that the number of comorbidities a patient has would correlate with the incidence of complications, but there was no relationship identified in this study. Is the number of comorbidities a real factor to influence complications? This finding may be due to our anesthesiologists’ stringent assessment of the patients preoperatively that eliminated the patients with a large number of comorbidities. In our study, wound infection occurred in 5 patients (4 %). This rate is similar to those reported in the literature contained between 0 and 15 % [1, 6, 8, 10, 12, 16, 17, 20, 24, 25] in the elderly patients and usually known in general population [14]. To our knowledge, the present study is the first to emphasize the prevalence of Gram negative bacteria in wound infection: E. coli (2 cases) and E. faecalis (2 cases), compared to one case of S. aureus. Whereas S. aureus is generally the cause of wound infections in younger patients following lumbar spine surgery, this series demonstrated that wound infections in patients over the age 80 years were more commonly caused by urinary tract or fecal organisms. This point is probably to compare in our study the prevalence of urinary retentions and urinary tract infections. Thus, we modified our preventive antibioprophylaxis and are most careful in the preoperative perineal preparation. In our study, dural tears occurred in 19 patients (15.7 %) and increased with previous surgery—33.3 versus 12.4 %—In the literature, this rate is very heterogeneous and contained between 0.003 and 16 % [1, 6, 8, 10, 12, 14, 17, 20, 21]. Some authors [2, 8, 12, 20, 23, 24] included
123
Eur J Orthop Surg Traumatol
patients who had undergone previous lumbar surgery, and they found no relation between dural tears and previous surgery. It is quite conceivable that previous surgery and the consequent dural adhesions could increase the risk of dural tears. Moreover, we found a link between major complications and dural tears. We hypothesize that this finding was due to the inability to drain the surgical site which favored the formation of an epidural hematoma. All of the patients with an epidural hematoma had a dural tear. Consequently, the avoidance of a dural tear is critically important to decrease postoperative morbidity specifically the operative time, blood loss and the development of a postoperative hematoma. Finally, although Olridge et al. [27] found an increasing rate of mortality in patients 80 years of age or older, our findings show that the incidence of mortality was not increased in the first year postoperatively. The main weakness of our study is its retrospective design. However, this study is continuous and all the patients had follow-up with a satisfactory time. The second weakness of this study was no comparison group which limited too affirmative conclusions. In conclusion, the morbidity of degenerative lumbar spine surgery in patients 80 years of age or older was high. Blood loss, operative time, instrumentation, previous surgery and dural tears increased significantly the morbidity. The lumbar spine surgery in patients 80 years of age or older must be decided very carefully and requires to inform the patient and his family of the high rate of morbidity. Instrumented fusion must be limited and more cautiously decided in older patients than in younger patients. Acknowledgments No funds were received in support of this work. No benefits in any form have been or will be received from a commercial party related directly or indirectly to the subject of this manuscript. Conflict of interest
None.
References 1. Best NM, Sasso RC (2007) Outpatient lumbar spine decompression in 233 patients 65 years of age or older. Spine 32:1135–1139 2. Raffo CS, Lauerman WC (2006) Predicting morbidity and mortality of lumbar spine arthrodesis in patients in their ninth decade. Spine 31:99–103 3. Sanderson PL, Wood PLR (1993) Surgery for lumbar spinal stenosis in old people. J Bone Joint Surg Br 75-B:393–397 4. Smith EB, Hanigan WC (1992) Surgical results and complications in elderly patients with benign lesions of the spinal canal. JAGS 40:867–870 5. Charlson ME, Pompei P, Ales KL et al (1987) A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Cron Dis 40:373–383 6. Carreon LY, Puno RM, Dimar JR et al (2003) Perioperative complications of posterior lumbar decompression and arthrodesis in older adults. J Bone Joint Surg Am 85-A:2089–2092
123
7. Tadokoro K, Miyamoto H, Sumi M et al (2005) The prognosis of conservative treatments for lumbar spinal stenosis. Analysis of patients over 70 years of age. Spine 30:2458–2463 8. Greenfield RT, Capen DA, Thomas JC et al (1998) Pedicle screw fixation for arthrodesis of the lumbosacral spine in the elderly. Spine 23:1470–1478 9. Kazunori Y, Takashi S, Yosihisa K et al (1996) Indication of fusion for lumbar spinal stenosis in elderly patients and its significance. Spine 21:242–248 10. Becker P, Bretschneider W, Tuschel A et al (2010) Life quality after instrumented lumbar fusion in the elderly. Spine 35:1478–1481 11. Galiano K, Obwegeser AA, Gabl MV et al (2005) Long-term outcome of laminectomy for spinal stenosis in octogenarians. Spine 30:332–335 12. Jo¨nsson B, Stro¨mqvist B (1994) Lumbar spine surgery in the elderly. Complications and surgical results. Spine 19:1431–1435 13. Rosen DS, O’Toole JE, Eichholz KM et al (2007) Minimally invasive lumbar spinal decompression in the elderly: outcomes of 50 patients aged 75 years and older. Neurosurg 60:503–510 14. Weinstein JN, Lurie JD, Tosteson TD et al (2007) Surgical versus nonsurgical treatment for lumbar degenerative spondylolisthesis. N Engl J Med 356:2257–2270 15. Zheng F, Sandhu HS, Cammisa FP et al (2001) Predictors of functional outcome in elderly patients undergoing posterior lumbar spine surgery. J Spinal Disord 14:518–521 16. Crawford CH, Smail J, Carreon LY et al (2011) Health-related quality of life after posterolateral lumbar arthrodesis in patients seventy-five years of age and older. Spine 36:1065–1068 17. Reindl R, Steffen T, Cohen L et al (2003) Elective lumbar spinal decompression in the elderly: Is it a high-risk operation? Can J Surg 46:43–46 18. Li G, Patil CG, Lad SP et al (2008) Effects of age and comorbidities on complication rates and adverse outcomes after lumbar laminectomy in elderly patients. Spine 33:1250–1255 19. Deyo RA, Cherkin DC, Loeser JD et al (1992) Morbidity and mortality in association with operations on the lumbar spine. J Bone Joint Surg Am 74-A:536–543 20. Ragab AA, Fye MA, Bohlman HH (2003) Surgery of the lumbar spine for spinal stenosis in 118 patients 70 years of age or older. Spine 28:348–353 21. Cassinelli EH, Eubanks J, Vogt M et al (2007) Risk factors for the development of perioperative complications in elderly patients undergoing lumbar decompression and arthrodesis for spinal stenosis. Spine 32:230–235 22. Deyo RA, Ciol MA, Cherkin DC et al (1993) Lumbar spinal fusion. A cohort study of complications, reoperations, and resource use in the Medicare population. Spine 18:1463–1470 23. Benz RJ, Ibrahim ZG, Afshar P et al (2001) Predicting complications in elderly patients undergoing lumbar decompression. Clin Orthop 384:116–121 24. Vitaz TW, Raque GH, Shields CB et al (1999) Surgical treatment of lumbar spinal stenosis in patients older than 75 years of age. J Neurosurg 91:181–185 25. Rodgers WB, Gerber EJ, Rodgers JA (2010) Lumbar fusion in octogenarians. The promise of minimally invasive surgery. Spine 35:S355–S360 26. De Thomasson E, Caux I, Guigand O, Terracher R, Mazel C (2009) Total hip arthroplasty for osteoarthritis in patients aged 80 years or older: influence of co-morbidities on final outcome. Orthop Traumatol Surg Res 95:249–253 27. Oldridge NB, Yuan Z, Stoll JE et al (1994) Lumbar spine surgery and mortality among medicare beneficiaries, 1986. Am J Public Health 84:1292–1298
