Reliability of the global alignment and proportion score in predicting mechanical complications after correction surgery of adult spinal deformities

English Keywords:Adult spinal deformity  Global alignment and proportion  Mechanical complication  Reliability

English Abstract:

  【Abstract】 Objectives: To determine the intra- and inter-observer reliabilities of global alignment and proportion(GAP) score in predicting the mechanical complications after correction surgery of adult spinal deformities(ASD). Methods: A total of 98 ASD patients undergone correction surgery between December 2014 and December 2016 with at least 2 years follow-up were retrospectively reviewed, including 31 male and 67 female subjects that aged 38.8±8.6(20-71) years on average. Patients′ age were documented and radiographic parameters including PI, SS, LL, LLL and GT were independently measured twice by 3 spine surgeons on postoperative lateral full-spine X-rays to determine the GAP scores and classifications. Patients were classified into 3 groups: proportion group with GAP scores between 0-2, moderate disproportion group with GAP scores between 3-6, severe disproportion group with GAP scores between 7-13. All mechanical complications including implant-related failures, PJK/PJF and DJK/DJF were documented. The intra- and inter-observer reliabilities were analyzed and the incidence of mechanical complications between the 3 groups was compared with Chi-square tests. The visual analogue scale(VAS), Oswestry disability index(ODI) and the MOS item short form health survey(SF-36) were compared among three groups. Results: Of all 588(98×3×2) measurements of GAP scores obtained, 192(32.7%) measurements were classified into the proportion group, 250(42.5%) into the moderate disproportion group, and 146(24.8%) into the severe disproportion group. The Kappa coefficients were 0.765 for inter-observer and 0.822 for intra-observer reliabilities. A total of 15 patients were identified to have mechanical complications during follow-up, and the incidences were 9.1%(3/33), 14.6%(6/41) and 25.0%(6/24) for proportion group, moderate disproportion group and severe disproportion group, respectively (χ2=2.74, P=0.254). At pre-operation, the VAS(P=0.668), ODI(P=0.167), SF-36 PCS(P=0.896) and SF-36 MCS(P=0.211) were not significantly different between the 3 groups. During the last follow up, the ODI(P=0.038) and SF-36 PCS(P=0.020) were significantly different between groups, no statistically significant difference was found in terms of VAS(P=0.729) and SF-36 MCS(P=0.277). Conclusions: The GAP scores have good to excellent intra- and inter-observer reliabilities in ASD patients undergoing correction surgery. The GAP scores are unable to predict the occurrence of mechanical complications during follow-up.