Kitchen elbow sign predicts surgical outcomes in adults with spinal deformity: a retrospective cohort study

Kitchen elbow sign (KE-Sign) is a skin abnormality on the extensor side of the elbow and forearm that is often observed in patients with adult spinal deformity (ASD). The significance of KE-Sign in surgical cases was investigated. Overall, 114 patients with ASD treated with long spinal fusion were reviewed and divided into KE-Sign positive and negative groups. The preoperative and 1-year follow-up evaluations included radiographic parameters [C7 sagittal vertical axis (SVA), pelvic incidence (PI) and lumbar lordosis (LL)], the Oswestry Disability Index (ODI), visual analogue scales (VASs) for low back pain, leg pain, and satisfaction, and Short Form 36 questionnaire (SF-36). Multi-regression analysis was performed to identify patient satisfaction predictors and improvement in the ODI as dependent variables and preoperative background factors as independent variables. Preoperative characteristics showed no significant difference between both groups. Improvement in the ODI and VAS for satisfaction were significantly superior in the KE-Sign positive group. In multiple regression analysis, KE-Sign and preoperative ODI were significantly associated with improvement in the ODI; age, KE-Sign, preoperative low back pain VAS, and leg pain VAS were significantly associated with satisfaction. KE-Sign can be a predictor of better surgical outcomes in ASD patients.

In most developed countries, the average human life expectancy has doubled over the past 200 years 1 . We are now living longer with a better quality of life than at any time in human history. However, disease-free lifespan has not increased as compared to the total lifespan, and an average of 16%-20% of our lives is spent with latelife morbidities 2,3 . The incidence of musculoskeletal disease, including spinal disorders, is increasing with the growing aging population. Adult spinal deformity (ASD) has gained significant attention in the last 10 years following improvements in diagnostic tools, classification schemes, and surgical techniques 4 . Furthermore, sagittal malalignment is widely known to correlate with clinical outcomes such as low back pain, and health-related quality of life (HRQOL) in patients with lumbar degenerative disorders, and especially ASD 5,6 . Non-surgical treatment is effective in ASD patients if they do not have significant radicular or mechanical back pain and have scoliosis less than 30° or subluxation less than 2 mm 7 . However, previous studies have reported that non-surgical treatment has not been successful for low back pain and radicular pain caused by ASD 8 . Some studies have demonstrated that surgical treatment is superior to non-surgical treatment in improving ASD patient outcomes 9,10 . With advances in minimally invasive surgery, represented by extreme lateral interbody fusion 11 , elderly people or high-risk patients who were previously inoperable can now undergo surgery. This implies that surgery for ASD is also expected to extend the healthy life span of elderly people. Although spinal fusion improves related symptoms and HRQOL in ASD patients, the loss of flexibility of the spinal column makes it difficult to perform daily activities that require trunk flexion 12 . Therefore, not all patients who receive surgical treatment for ASD obtain satisfactory results. Identifying preoperative factors that can predict surgical outcomes and patient satisfaction is very important in treating ASD. Many clinicians have reported about predictors of complications as well as mortality following ASD surgery 13,14 . However, limited information exists on preoperative predictors of surgical outcomes or patient satisfaction. In 2012, Miyamoto et al. 15 1) on the extensor surface of the elbows and forearms of patients with intractable low back pain in the standing position. They reported that the KE-Sign is a skin change that runs from the elbow to the forearm in patients with intractable low back pain in the standing position, and is caused by supporting oneself using the elbows in a standing position during housework (e.g., during kitchen work). The KE-Sign is not specific to ASD patients, but may act as a surrogate marker of maintaining and continuing an independent lifestyle even with intractable low back pain due to sagittal plane imbalance if the population is limited to patients with ASD. Therefore, we hypothesized that the KE-Sign correlates with HRQOL and can be used to predict surgical outcomes or patient satisfaction in ASD. This study aimed to investigate the significance of KE-Sign in surgical cases of ASD.

Materials and methods
We hypothesized that KE-Sign can be used to predict surgical outcomes and patient satisfaction in ASD. This study was reviewed and approved by the local ethics committee (Research Ethics Committee of Wakayama Medical University: approval number 2943) and has been performed in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki and its later amendments. Informed consent was obtained from all participants.
Between January 2015 and August 2019, all consecutive patients who had (1) symptomatic ASD and (2) received long corrective fusion from the thorax to the pelvis, were enrolled in this study. We defined ASD as C7 sagittal vertical axis (SVA) > 50 mm and pelvic incidence-lumbar lordosis (PI-LL) > 10° in accordance with the SRS-Schwab Classification 16 . The exclusion criteria were as follows: (1) history of spine surgery, (2) bedridden status due to pre-existing problems, (3) current severe infection, or concurrent acute fracture. The initial criteria for inclusion were met by 118 cases. Three patients who had a history of cervical spinal surgery were excluded, and after excluding 1 patient because of cerebral infarction, the remaining 114 patients were included in the final analysis (Fig. 2). The KE-Sign was evaluated at the outpatient clinic before surgery in all cases. Of the 3 parameters: (1) intractable low back pain in standing position, (2) pigmentation of dorsal forearm by visual inspection and (3) rough skin on the dorsal forearm by palpation, KE-Sign positive was defined as always satisfying (1) and satisfying any one of (2) and (3).
Prior to this study, 50 patients were examined by two orthopedic surgeons to determine inter-rater reliability of this definition by using Cohen's Kappa test. The inter-rater reliability was Kappa 0.96. Clinical outcome measurements. Among indicators used to evaluate surgical outcomes of ASD, the Oswestry Disability Index (ODI) is the most commonly used 17 . The ODI was determined for every patient in our study 18 . It has 10 subclasses that measure pain intensity, personal care, lifting, walking, sitting, standing, sleeping, social life, sex life, and traveling. For each subclass, scores range from 0 (best measured health) to 5 (worst measured health). In this study, we used the Japanese version of ODI 19 that excluded components related to sex life because limited information is expected for this component because of the Japanese cultural background; thus, the final score may be calculated as follows: (total score/(5 × 9 questions excluding "sex life")) × 100%). ODI change score was calculated as preoperative ODI%-12-month postoperative ODI%. Moreover, the achievement rate of minimum clinically important difference (MCID) of ODI change score was evaluated with reference to the result of the previous study 20 . In order to evaluate patient satisfaction, we used the satisfaction visual analog scale (VAS). A 100-mm VAS has word descriptors "very dissatisfied" to "very satisfied" on the left and right, www.nature.com/scientificreports/ respectively. Subsequently, all scales were measured from the left to right to obtain the numeric values as the scores for patients' satisfaction 21 .
We evaluated age, sex, body mass index (BMI), and presence or absence of KE-Sign as baseline characteristics of the patients; C7 SVA as an indicator of the whole spine alignment; PI-LL as an indicator of lumbar pelvis sagittal plane alignment; and ODI, lumbar VAS (low back pain and leg pain), and SF-36 as patient-based items. The SF-36 Physical Component Summary (PCS) and Mental Component Summary (MCS) scores were obtained using reported algorithms from a previous questionnaire 22 . KE-Sign was evaluated only before surgery while satisfaction VAS was evaluated only after surgery. All other parameters were evaluated before and 1 year after surgery. All results were collated and analyzed independently of the operating surgeon.

Statistical analysis.
To compare data between the KE-Sign positive and KE-Sign negative groups, Fisher's exact test/chi square test was used for proportional variables. Analysis of variance (ANOVA) was performed for continuous variables. First, baseline characteristics and postoperative outcomes at 1 year after surgery were compared between the two groups. Then, for intragroup comparisons between baseline and one-year post-surgery, a paired t-test was used. The intervention effect (the change of each parameter including interaction) was also compared by a multivariate analysis of variance (MANOVA). Finally, multi-regression analysis was performed to identify predictors of patient satisfaction and improvement in ODI as dependent variables, and sex, age, BMI, presence or absence of KE-Sign, preoperative C7 SVA, PI-LL, ODI (%), lumbar VAS, and component summary scales of SF-36 as independent variables. All statistical analyses were performed using JMP Pro version 14 (SAS Inc., Cary, NC, USA), with the level of significance set at P < 0.05.

Results
Participant selection and comparison of preoperative items between KE-Sign positive group and KE-Sign negative group. One hundred and eighteen cases met the initial criteria for inclusion. Three patients who had a history of cervical spinal surgery were excluded, and after excluding 1 patient because of cerebral infarction, 114 patients were included in the final analysis (Fig. 2). The average age of the 114 patients  Comparison of the one-year postoperative outcomes between KE-Sign positive group and KE-Sign negative group. All evaluated items including C7 SVA, PI-LL, VASs of low back pain and leg pain, SF-36 component summary scales, and ODI score were significantly improved from the baseline in both groups. Postoperative C7 SVA, PI-LL, VASs of low back pain and leg pain, and SF-36 component summary scales were not significantly different between the two groups (Table 2). However, the ODI score was significantly different between the two groups at 1 year after surgery. Furthermore, the MCID achievement rate of ODI score was significantly higher in the KE-Sign positive group (P = 0.003) ( Table 2). Patient satisfaction VAS was significantly higher in the KE-Sign positive group (P < 0.001) ( Table 2). Based on the above, the KE-Sign positive group showed better surgical outcomes and patient satisfaction than the KE-Sign negative group.
Intergroup comparison of the change of each parameter. Changes in each parameter were compared between the KE-Sign positive group and KE-Sign negative group (Table 3). The effect of time (i.e., surgical intervention) was significant for C7 SVA, PI-LL, lumbar VASs, SF-36 component summary scales, and ODI score. Moreover, ODI scores showed significant differences in group, and interaction. www.nature.com/scientificreports/

Predictive factors for improving of ODI. Multiple regression analysis showed that KE-Sign positivity
and preoperative ODI score were the significant factors predicting the ODI improvement. The standard partial regression coefficients were 0.30 and 0.61, respectively (Table 4).

Predictive factors for postoperative patient satisfaction. Again, multiple regression analysis
showed that age, KE-Sign positivity, and preoperative VASs of low back pain and leg pain were the significant factors in predicting the patient's satisfaction at 1-year post-surgery. The standard partial regression coefficients were 0.29, 0.43, − 0.23, and 0.22, respectively (Table 5). Thus, older age, high leg pain intensity, and presence of KE-Sign predicted high satisfaction while high low back pain intensity predicted low satisfaction.

Discussion
The KE-Sign is often observed in patients who live an independent life. Until we obtained the results of this study, we believed that the KE-Sign correlates with HRQOL. Therefore, we hypothesized that the KE-Sign can be used to predict surgical outcomes or patient satisfaction in ASD. However, there are no reports investigating the relationship between the KE-Sign and ASD surgical outcomes and patient satisfaction. Therefore, we compared preoperative and postoperative parameters between the two groups, and performed multi-regression analysis for patient satisfaction and improvement in ODI, in order to investigate the significance of the KE-Sign in surgical cases of ASD. In this study, none of the preoperative characteristics of the patients were significantly different between the two groups. On the other hand, at the time of evaluation 1 year after surgery, the KE-Sign positive group showed significantly better surgical outcomes and patient satisfaction than the KE-Sign negative group. Moreover, in multi-regression analysis, KE-Sign positivity was significantly associated with both improvement in ODI and high satisfaction. These results suggest that the presence of the KE-Sign may be a predictor of high satisfaction and good surgical outcome. In other words, if KE-Sign exists preoperatively, good clinical results may be expected. We do not believe that the KE-Sign is specific to ASD patients or superior to other predictive It is reasonable to believe that patients with both indoor and outdoor ADL restrictions that are improved after surgery would be more satisfied than patients that only experience improvement in outdoor ADL restrictions. Accordingly, we believe that KE-Sign ( +) can be a surrogate marker of good surgical indication for ASD and patient satisfaction. There was no significant difference between the KE-Sign ( +) and KE-Sign (−) groups in preoperative evaluation results, including pain intensity, SF-36 component summary scales, and ODI score. However, these measures do not sufficiently evaluate the ability to perform household chores. Therefore, we believe that KE-Sign may act as a surrogate marker of increased postoperative satisfaction with respect to the preoperative inability to perform household chores. Previous studies, which examined preoperative predictors of surgical outcomes after spinal long fusion surgery, have not shown a significant correlation between age, BMI, spine fusion range, operative time, blood volume, and preoperative PI-LL mismatch 13,14 . In some reports, patients with high preoperative ODI scores had significant improvement in ODI after surgery 23 . However, other reports have shown that patients with high preoperative ODI score have poor surgical outcome 24 . In addition, since patient satisfaction is affected by the patient's mental and physical health, the physician-patient relationship, incidence of perioperative complications, quality of the perioperative nursing care, and the patient's expectations, among others, it is difficult to determine the predictors of patient satisfaction 25 . Only limited information exists on preoperative predictors of surgical outcomes or patient satisfaction. This study is the first report investigating the relationship between KE-Sign and surgical outcomes of ASD and patient satisfaction in surgical cases of ASD. This study suggests that KE-Sign is a preoperative factor that could predict both a high postoperative satisfaction level and good surgical outcomes.
Nevertheless, this study had some limitations. First, this study follows a retrospective design; thus, further studies will be needed to compare this patient group as the size of this patient population increases. Second, this study only addressed the effect of preoperative patient and surgical characteristics on 12-month postoperative patient satisfaction and surgical outcome, but did not address the effect of postoperative factors on long-term satisfaction and outcomes. Third, a globally standardized diagnostic criterion of KE-Sign is not established. Since KE-Sign is a new concept, there are still many unclear points about the pathology and significance of KE-Sign with ASD. There is a possibility that the results of this study cannot be generalized to other countries because the target population was limited to Japanese patients. However, since the Japanese lifestyle has already been undergoing westernization for decades, we believe that the results of this study will be reproduced in other countries as well. However, further research is needed to determine whether the results of this study can be adapted to other countries in which lifestyle and social welfare system are different from Japan. Fourth, the possibility that multiple factors may have confounded the relationship between the preoperative KE-Sign and postoperative results cannot be completely ruled out. We also did not report radiographic parameters, such as T1PA, GSA, C2-7SVA, C2-7 angle, T1-T12 angle, T11-L2 angle, and hip and knee angles, in this study because we intended to focus on the relationship between the KE-Sign and surgical outcomes, mainly in terms of HRQOL. The results of this study indicated that the KE-Sign ( +) predicted better improvement of ODI scores in ASD patients. However, HRQOL measures other than the ODI, such as the SRS-22, were not routinely evaluated during the study period. Future