Annabel Dekker1, Haemish A Crawford1, N Susan Stott1. 1. A Dekker, H. A. Crawford, N. S. Stott, Department of Paediatric Orthopaedics, Starship Children's Hospital, Auckland, New Zealand.
Abstract
BACKGROUND: Surgery for spinal deformity in patients with cerebral palsy is reported to have high perioperative complication rates. However, minor complications are not generally reported and the influence of the varied severity of complications on length of stay is not known. Understanding the risk factors for both minor and major perioperative complications and their effect on length of stay is important information for clinicians who seek to improve care for this group of children. QUESTIONS/PURPOSES: (1) What is the prevalence of postoperative complications in the first 30 days after surgery for spinal deformity in a New Zealand national cohort of children with cerebral palsy using the Clavien-Dindo classification? (2) What are the patient and operative predictive risk factors for minor and major perioperative complications? (3) What is the effect of year of operation on risk of minor and major perioperative complications? (4) What is the effect of perioperative complications on length of stay? METHODS: We conducted a retrospective cohort study, identifying all children in New Zealand with a confirmed diagnosis of cerebral palsy who had surgery for a spinal deformity from January 1997 to January 2018. Two hundred-three patients with cerebral palsy (102 boys) were surgically treated for a spinal deformity, at a mean age of 14 ± 3 years, at one of three centers in New Zealand. Six children had Gross Motor Function Classification System Level II or III, 66 had Gross Motor Function Classification System Level IV, and 131 had Gross Motor Function Classification System Level V. Thirty-day perioperative complications were extracted from the patients' health records and classified according to the Clavien-Dindo system. Univariate and multivariate analyses were performed to identify patient and operative risk factors for complications, and the effect on length of stay. RESULTS: In all, 85% of patients experienced at least one perioperative complication. There were 300 Clavien-Dindo Grade I complications in 141 patients, 156 Clavien-Dindo Grade II complications in 102 patients, 25 Clavien-Dindo Grade III complications in 22 patients, 29 Clavien-Dindo Grade IV complications in 28 patients, and one Clavien-Dindo Grade V complication (death; 0.5%). Univariate analysis showed that multiple independent factors, Gross Motor Function Classification System Level V ability (odds ratio 2.13 [95% confidence interval 1.15 to 3.95]; p = 0.02), seizure disorder (OR 2.27 [95% CI 1.20 to 4.32]; p < 0.01), preoperative Cobb angle of greater than 70° (OR 2.40 [95% CI 1.20 to 4.78]; p < 0.01), and anterior approach to surgery (OR 3.29 [95% CI 1.21 to 8.90]; p = 0.02), were associated with Grade I complications but, of these factors, only the presence of a seizure disorder (OR 2.27 [95% CI 1.20 to 4.32]; p < 0.01) was associated with Grade I complications on multivariate analysis. Previous recurrent respiratory infections predicted an increased risk of Clavien-Dindo Grade II complications (OR 3.6 [95% CI 1.81 to 7.0]; p = 0.03). The presence of a feeding gastrostomy was associated with an increased risk of Clavien-Dindo Grade IV complications (OR 2.6 [95% CI 1.19 to 5.87]; p = 0.02). The year of operation did not influence the frequency of any grade of complication, but the presence of any complication led to an increased length of stay. CONCLUSION: Overall, 85% of patients with cerebral palsy had at least one complication after spinal deformity surgery and 25% had major complications (Grades III, IV, and V), with proportionate increases in the postoperative length of stay. Patient-specific factors aid in the identification of complication risk. LEVEL OF EVIDENCE: Level II, prognostic study.
BACKGROUND: Surgery for spinal deformity in patients with cerebral palsy is reported to have high perioperative complication rates. However, minor complications are not generally reported and the influence of the varied severity of complications on length of stay is not known. Understanding the risk factors for both minor and major perioperative complications and their effect on length of stay is important information for clinicians who seek to improve care for this group of children. QUESTIONS/PURPOSES: (1) What is the prevalence of postoperative complications in the first 30 days after surgery for spinal deformity in a New Zealand national cohort of children with cerebral palsy using the Clavien-Dindo classification? (2) What are the patient and operative predictive risk factors for minor and major perioperative complications? (3) What is the effect of year of operation on risk of minor and major perioperative complications? (4) What is the effect of perioperative complications on length of stay? METHODS: We conducted a retrospective cohort study, identifying all children in New Zealand with a confirmed diagnosis of cerebral palsy who had surgery for a spinal deformity from January 1997 to January 2018. Two hundred-three patients with cerebral palsy (102 boys) were surgically treated for a spinal deformity, at a mean age of 14 ± 3 years, at one of three centers in New Zealand. Six children had Gross Motor Function Classification System Level II or III, 66 had Gross Motor Function Classification System Level IV, and 131 had Gross Motor Function Classification System Level V. Thirty-day perioperative complications were extracted from the patients' health records and classified according to the Clavien-Dindo system. Univariate and multivariate analyses were performed to identify patient and operative risk factors for complications, and the effect on length of stay. RESULTS: In all, 85% of patients experienced at least one perioperative complication. There were 300 Clavien-Dindo Grade I complications in 141 patients, 156 Clavien-Dindo Grade II complications in 102 patients, 25 Clavien-Dindo Grade III complications in 22 patients, 29 Clavien-Dindo Grade IV complications in 28 patients, and one Clavien-Dindo Grade V complication (death; 0.5%). Univariate analysis showed that multiple independent factors, Gross Motor Function Classification System Level V ability (odds ratio 2.13 [95% confidence interval 1.15 to 3.95]; p = 0.02), seizure disorder (OR 2.27 [95% CI 1.20 to 4.32]; p < 0.01), preoperative Cobb angle of greater than 70° (OR 2.40 [95% CI 1.20 to 4.78]; p < 0.01), and anterior approach to surgery (OR 3.29 [95% CI 1.21 to 8.90]; p = 0.02), were associated with Grade I complications but, of these factors, only the presence of a seizure disorder (OR 2.27 [95% CI 1.20 to 4.32]; p < 0.01) was associated with Grade I complications on multivariate analysis. Previous recurrent respiratory infections predicted an increased risk of Clavien-Dindo Grade II complications (OR 3.6 [95% CI 1.81 to 7.0]; p = 0.03). The presence of a feeding gastrostomy was associated with an increased risk of Clavien-Dindo Grade IV complications (OR 2.6 [95% CI 1.19 to 5.87]; p = 0.02). The year of operation did not influence the frequency of any grade of complication, but the presence of any complication led to an increased length of stay. CONCLUSION: Overall, 85% of patients with cerebral palsy had at least one complication after spinal deformity surgery and 25% had major complications (Grades III, IV, and V), with proportionate increases in the postoperative length of stay. Patient-specific factors aid in the identification of complication risk. LEVEL OF EVIDENCE: Level II, prognostic study.
Authors: Daniel L Master; Jochen P Son-Hing; Connie Poe-Kochert; Douglas G Armstrong; George H Thompson Journal: Spine (Phila Pa 1976) Date: 2011-04-01 Impact factor: 3.468
Authors: Rachel Toovey; Adrienne Harvey; Michael Johnson; Louise Baker; Katrina Williams Journal: Dev Med Child Neurol Date: 2017-03-06 Impact factor: 5.449
Authors: E Ascani; P Bartolozzi; C A Logroscino; P G Marchetti; A Ponte; R Savini; F Travaglini; R Binazzi; M Di Silvestre Journal: Spine (Phila Pa 1976) Date: 1986-10 Impact factor: 3.468
Authors: Emily R Dodwell; Rubini Pathy; Roger F Widmann; Daniel W Green; David M Scher; John S Blanco; Shevaun M Doyle; Aaron Daluiski; Ernest L Sink Journal: JB JS Open Access Date: 2018-10-23