Kyphoplasty for vertebral compression fractures: the blue zone effect Authors: David Shin, Elijah Haynal, Rekha Isaac, Graydon Dietrich, Hyun Ah Park, Daniel Im, Jeffrey Lubisich, Brandon Edelbach, Othman Bin-Alamer, Rasha Elbadry, Peter Zaki, Namath Hussain Published: 2026 ABSTRACT BackgroundKyphoplasty effectively treats vertebral compression fractures (VCFs) but carries risks of perioperative complications and readmissions. Blue zones, regions characterized by longevity and reduced chronic disease, provide a unique opportunity to examine how lifestyle factors may influence surgical outcomes. This study investigated how patient demographics, health status, surgical factors, and geographic location, particularly residence in a blue zone, affected kyphoplasty outcomes.MethodsA retrospective cohort review of 372 patients undergoing kyphoplasty for VCFs from January 2009 to May 2025 was conducted at a single, tertiary care institution. Patients were divided into blue zone or non-blue zone by postal codes. 1:3 propensity score matching (PSM) was used to minimize bias and multivariable logistic regression analyzed outcomes. A 10-year Kaplan-Meier survival curve was formulated utilizing a log-rank test.ResultsA total of 190 patients (175 non-blue zone and 15 blue zone patients) were available after exclusion. In the unmatched cohort, tobacco usage was associated with higher 30-day readmissions. Increased length of stay (LOS) was associated with higher American Society of Anesthesiologists (ASA) scores, older age, diabetes, mechanical expansion, and greater estimated blood loss (EBL). Thoracolumbar involvement increased neurological complications, while a bilateral approach reduced incidence. Blue zone patients had significantly lower alcohol (0% vs. 34–64%) and tobacco use (7% vs. 43–47%) than non-blue zone patients in both unmatched and 1:3 PSM cohorts. No other significant differences in outcomes were observed.ConclusionsOur findings reinforce the significance of traditional risk factors including age, ASA score, diabetes, and tobacco use, as key indicators of kyphoplasty outcomes. Blue zone kyphoplasty patients demonstrated markedly healthier lifestyles, such as reduced alcohol and tobacco use. The present small-sized matched analyses demonstrated that while lifestyle and environmental factors may delay onset of chronic disease and VCF occurrence, no detectable association was identified between blue zone residence and kyphoplasty outcomes. Highlight box Key findings • Blue zone patients had significantly lower alcohol (0% vs. 34–64%) and tobacco use (7% vs. 43–47%) than non-blue zone patients in both unmatched and 1:3 propensity score matched cohorts. No detectable association was identified between blue zone residence and kyphoplasty outcomes. In the unmatched cohort, tobacco usage was associated with higher 30-day readmissions. Increased length of stay was associated with higher American Society of Anesthesiologists (ASA) scores, older age, diabetes, mechanical expansion, and greater estimated blood loss. What is known and what is new? • Blue zones are regions characterized by exceptional longevity and low incidence of chronic disease such as osteoporosis. They have been theorized to arise from a combination of dietary habits, physical activity, social engagement, and genetic factors, showcasing delayed onset of frailty and osteoporotic hip fractures by 12 years, and fragility vertebral compression fractures by about 10 years. However, the degree to which blue zone characteristics may affect perioperative surgical outcomes remains poorly characterized in the literature. • Despite more favorable lifestyle habits, our small sized-matched analyses demonstrate a lack of detectable association between blue zone status and kyphoplasty outcomes. What is the implication, and what should change now? • Traditional risk factors including age, ASA score, diabetes, and tobacco remain the primary indicators of kyphoplasty outcomes. Geographic residence as a proxy for blue zone lifestyle may be an oversimplification without further validation. The small blue zone cohort warrants caution, with a lack of detectable association identified between blue zone residence and kyphoplasty outcomes. Introduction Vertebral compression fractures (VCFs) are among the most common fragility fractures and are associated with significant morbidity, mortality, and healthcare costs (1). Often linked to aging, chronic disease, and osteoporosis, VCFs may profoundly impact on an individual’s quality of life, causing pain, disability, and reduced functional capacity (2). Cement augmentation procedures such as kyphoplasty have emerged as effective interventions for managing symptomatic VCFs. By stabilizing the fracture and preventing further deformity, kyphoplasty has shown efficacy in reducing pain and improving functional ability (3,4). However, despite clinical benefits, kyphoplasty has also been linked to perioperative complications and variable outcomes influenced by patient- and procedure-related factors. Given the established relationship between environmental and lifestyle factors, chronic disease, and surgical risk, the degree to which blue zone characteristics may affect perioperative surgical outcomes remains poorly characterized in the literature. “Blue zones”, regions characterized by exceptional longevity and low incidence of chronic disease such as osteoporosis, have been theorized to arise from a combination of dietary habits, physical activity, social engagement, and genetic factors (5,6). As such, blue zones offer a unique opportunity to explore the impact of lifestyle behaviors and environmental influences on surgical outcomes and postoperative complications. To address these needs, this study aimed to investigate how patient demographics, health status, surgical factors, and geographic location, particularly residence in a blue zone, affected outcomes following kyphoplasty for VCFs. We present this article in accordance with the STROBE reporting checklist (available at https://jss.amegroups.com/article/view/10.21037/jss-2026-0095/rc). Methods Study design and population This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Loma Linda University (approval No. 5240326). Informed consent was waived due to the nature of this retrospective study. Following approval, a retrospective cohort review of 372 patients who underwent kyphoplasty for VCFs from January 2009 to May 2025 was conducted at a single tertiary care institution in Loma Linda. Patients with primary and metastatic spinal malignancies, instrumental or surgical revision, high-energy mechanism and/or pathologic fractures, vertebroplasty, and scoliosis surgery were excluded. After implementation of exclusion criteria and removal of duplicates, complete data were available for 190 patients (175 non-blue zone and 15 blue zone patients). Variable selection Blue zone status was determined via the patient’s recorded address and postal code at the time of surgery. Although Loma Linda has been identified to be a blue zone in prior literature, the geographic boundaries have not been specifically defined. Publicly available municipal and postal boundary resources were utilized to identify postal codes corresponding to Loma Linda. Patients with postal codes corresponding to Loma Linda residence (e.g., 92354 and 92373) were subsequently designated as blue zone residents. Duration of residence was not formally available within the database, but duration of residence was assessed beyond the isolated time of surgery through extended chart review, including verification of continued residence within Loma Linda from the earliest documented encounters through the most recent encounters available in the medical record. Key variables were collected via electronic medical records including demographics, which consisted of patient age, sex, race and ethnicity, insurance status, and religion. Patient health metrics and chronic medical conditions recorded consisted of body mass index (BMI), preoperative ambulatory status and patient-reported pain scores, diabetes, osteoporosis, chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), hypertension, chronic kidney disease (CKD), current or prior non-spinal malignancy, as well as current or prior alcohol, tobacco, and drug usage. Surgical modalities Surgical modalities collected included surgical service, vertebral level and number of vertebrae treated, degree of vertebral height loss, American Society of Anesthesiologists (ASA) score, estimated blood loss (EBL), cement type/manufacturer, cement volume, surgical approach, balloon or mechanical inflation, and total intraoperative fluoroscopy usage. Table 1 showcases a full breakdown of collected variables. Table 1Collected variables and number of patients (n=190)VariablesDataSex Male50 Female140Blue zone status Blue zone15 Non-blue zone175Race/ethnicity White133 Black6 Hispanic35 Asian13 Native-American3Insurance status Public116 Private70 N/A4Religion Christian130 SDA26 Muslim3 Jewish2 Non-denominational12 Atheist6 Other11Diabetes Yes35 No155Osteoporosis Yes149 No41Hypertension Yes134 No56COPD Yes27 No163CHF Yes29 No161CKD None160 Stage 1 (GFR >90)2 Stage 2 (GFR 60–89)5 Stage 3 (GFR 30–59)17 Stage 4 (GFR 15–29)3 Stage 5 (GFR <15)3Non-spinal malignancy None131 Current24 Previous35Alcohol usage None131 Current40 Previous19Tobacco usage None106 Current19 Previous65Drug usage None166 Current15 Previous9ASA score 10 250 3132 48 50Pre-operative ambulatory status Independent68 Assisted (walker, cane, etc.)87 Non-ambulatory (bed/wheelchair bound, etc.)35Post-operative ambulatory status Independent92 Assisted (walker, cane, etc.)81 Non-ambulatory (bed/wheelchair bound, etc.)17Vertebral region treated Cervical0 Thoracic68 Lumbar86 Sacral2 Thoracolumbar33 Lumbosacral1Surgical complications (within 30 days) None150 Bleeding3 New neurologic deficit (paresthesia, altered mental status, and paralysis)19 Dural tear/CSF leak2 Infection6 Wound dehiscence1 New vertebral fracture7 Cement leakage2Related readmission (within 30 days) Yes21 No169Revision surgery (within 2 years) Yes169 No21Mortality Yes21 No169Surgical approach Unilateral30 Bilateral160Vertebral body augmentation technique Balloon141 Mechanical49Cement type/manufacturer Confidence Spinal Cement System (DePuy Synthes)105 CX01A/B Xpede Bone Cement (Medtronic)63 VertaPlex Cement (Stryker)22Surgical service Neurosurgery67 Orthopedic surgery123Age (years)69.6±12.3BMI (kg/m2)27.4±5.7DEXA t-score of lumbar spine−1.48±1.66LOS (days)1.2±2.28Pre-operative pain score7.01±2.2Post-operative pain score4.75±2.84Vertebral levels treated1.75Degree of height loss (%)44.89EBL (mL)18.11Cement volume (mL or cc)7.63Fluoroscopy time (seconds)124.09Data are presented as number of patients, mean ± SD, or mean. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CSF, cerebrospinal fluid; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; GFR, glomerular filtration rate; LOS, length of stay; N/A, not available; SD, standard deviation; SDA, Seventh Day Adventist. Collected variables and number of patients (n=190) Data are presented as number of patients, mean ± SD, or mean. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CSF, cerebrospinal fluid; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; GFR, glomerular filtration rate; LOS, length of stay; N/A, not available; SD, standard deviation; SDA, Seventh Day Adventist. Outcome measures Outcomes were assessed utilizing postoperative ambulatory status and patient-reported pain scores, hospital length of stay (LOS), postoperative complications, 30-day readmission, 2-year reoperation, and mortality. Analysis was first conducted on the entire cohort, then subsequently between the blue zone and non-blue zone groups. Statistical analysis Data analysis was conducted using R version 4.4.3 (R Core Team, 2025, Vienna, Austria). All outcome variables were analyzed using multivariable linear or logistic regression models as appropriate, with penalized regression employed for unstable predictors. Patient data were further divided into blue zone and non-blue zone groups based on postal codes, and 1:3 propensity score matching (PSM) was used to minimize bias. Due to a small sample size, standardized mean differences remained substantially unbalanced after PSM, and it should be interpreted conservatively. A Kaplan-Meier 10-year survival curve was formulated, and a log-rank test was applied. The significance level for all analyses was set at P=0.05. Results Total cohort characteristics Regarding the total unmatched cohort (n=190), the mean age was 69.6±12.3 years and the mean BMI was 27.4±5.7 kg/m2. The cohort included 50 males and 140 females, with 70.0% (n=133) identifying as White and 13.7% identifying as Seventh Day Adventist (SDA) (n=26). Common comorbidities included osteoporosis (78.4%, n=149), hypertension (70.5%, n=134), diabetes (18.4%, n=35), COPD (14.2%, n=27), and CHF (15.3%, n=29). Mean LOS was 1.2±2.28 days. Mean preoperative pain score was 7.01±2.2 and mean postoperative pain score was 4.75±2.84. Rates for related readmission within 30 days, reoperations within 2 years, and mortality were all 11% (n=21) (Table 1). In the total cohort, tobacco usage [P=0.008; odds ratio (OR) =8.753; 95% confidence interval (CI): 1.7–97.3] was associated with increased 30-day readmission rates. Age (P=0.01; OR =1.016; 95% CI: 1–1.03), diabetes (P=0.006; OR =1.756; 95% CI: 1.16–2.66), ASA scores (P<0.01; OR =3.7; 95% CI: 2.22–6.59), EBL (mL) (P=0.01; OR =1.003; 95% CI: 1.001–1.007), and the use of mechanical reduction (P=0.003; OR =1.87; 95% CI: 1.24–2.95) were associated with an increased LOS. The involvement of thoracolumbar levels (P=0.047; OR =7.12; 95% CI: 1.03–73.8) was associated with increased incidence of post-op neurological complications, and a bilateral kyphoplasty (P=0.03; OR =0.18; 95% CI: 0.03–0.86) was associated with reduced incidence of post-op neurological complications. Table 2 showcases a breakdown of all cohort outcomes pre-PSM. Table 3 showcases cohort comparisons pre-PSM. Table 2Cohort outcomes before PSMVariablesLOSNew neurologic deficit†30-day readmission†2-year reoperation†VAS pain score improvementAmbulatory improvement‡Mean ratio (95% CI)P valueaOR (95% CI)P valueaOR (95% CI)P valueaOR (95% CI)P valueβ (95% CI)P valueβ (95% CI)P valueBlue zone (vs. non-blue zone)1.66 (0.945, 3.031)0.061.156 (0.158, 8.006)0.882.147 (0.275, 17.302)0.455.673 (0.557, 213.221)0.14−0.414 (−2.304, 1.476)0.670.179 (−0.151, 0.51)0.29Age (per year)1.016 (1.003, 1.028)0.010.971 (0.921, 1.019)0.231.03 (0.963, 1.116)0.410.975 (0.872, 1.061)0.530.03 (−0.012, 0.073)0.160.003 (−0.005, 0.01)0.49Female (vs. male)1.314 (0.937, 1.824)0.103.03 (0.643, 34.844)0.172.198 (0.295, 23.015)0.450.878 (0.09, 23.19)0.91−0.105 (−1.232, 1.022)0.86−0.224 (−0.424, −0.023)0.03Orthopedic service (vs. neurosurgery)0.909 (0.61, 1.362)0.610.418 (0.094, 1.765)0.230.5 (0.044, 3.105)0.471.449 (0.138, 21.793)0.72−0.413 (−1.666, 0.84)0.520.181 (−0.039, 0.401)0.11Race Black (vs. White)2.667 (1.114, 7.367)0.021.877 (0, 123.25)0.780.088 (0, 18.109)0.341.594 (0.003, 71.012)0.790.512 (−2.306, 3.33)0.720.307 (−0.196, 0.809)0.23 Hispanic (vs. White)0.908 (0.608, 1.38)0.642.365 (0.476, 11.751)0.282.707 (0.406, 22.57)0.300.451 (0.006, 6.902)0.580.92 (−0.472, 2.312)0.20−0.006 (−0.252, 0.241)0.97 Asian (vs. White)1.377 (0.79, 2.643)0.270.373 (0.024, 2.714)0.350.311 (0.011, 5.045)0.430.437 (0, 6.491)0.610.922 (−0.987, 2.831)0.350.012 (−0.331, 0.355)0.95 Native (vs. White)1.273 (0.466, 4.358)0.663.638 (0.103, 88.055)0.4227.952 (0.051, 16091.529)0.240.414 (0.001, 81.376)0.670.065 (−3.597, 3.728)0.97−0.024 (−0.678, 0.631)0.94BMI (per kg/m2)0.986 (0.958, 1.016)0.310.974 (0.851, 1.096)0.660.933 (0.769, 1.087)0.410.991 (0.822, 1.126)0.90−0.03 (−0.124, 0.064)0.54−0.01 −0.027, 0.006)0.22Diabetes1.756 (1.158, 2.659)0.0062.163 (0.41, 11.637)0.355.75 (0.69, 70.041)0.112.381 (0.303, 23.837)0.380.843 (−0.528, 2.214)0.23−0.194 (−0.436, 0.049)0.12Osteoporosis0.643 (0.44, 0.929)0.0161.287 (0.291, 8.642)0.753.801 (0.478, 93.84)0.231.286 (0.125, 61.307)0.840.264 (−0.944, 1.471)0.670.005 (−0.212, 0.221)0.97COPD1.35 (0.881, 2.106)0.1491.495 (0.257, 7.664)0.620.489 (0.019, 4.498)0.571.605 (0.158, 12.307)0.640.721 (−0.646, 2.088)0.300.051 (−0.194, 0.296)0.68CHF1.257 (0.834, 1.908)0.2531.102 (0.188, 5.727)0.912.526 (0.332, 19.644)0.350.834 (0.078, 6.045)0.860.802 (−0.529, 2.134)0.24−0.059 (−0.297, 0.179)0.63Hypertension1.048 (0.742, 1.467)0.7651.094 (0.309, 4.879)0.893.927 (0.692, 63.274)0.130.683 (0.085, 6.187)0.69−0.593 (−1.664, 0.478)0.28−0.089 (−0.278, 0.099)0.35CKD0.698 (0.261, 1.96)0.4520.2 (0.001, 9.235)0.461.143 (0.04, 38.049)0.941.401 (0.003, 668.967)0.882.072 (−1.094, 5.237)0.200.344 (−0.221, 0.908)0.24Non-spine malignancy Prior1.242 (0.837, 1.853)0.251.703 (0.358, 7.029)0.480.596 (0.068, 3.326)0.581.296 (0.028, 32.849)0.83−0.076 (−1.355, 1.203)0.91−0.167 (−0.393, 0.059)0.15 Current0.719 (0.468, 1.129)0.120.255 (0.008, 1.743)0.181.432 (0.186, 10.132)0.723.486 (0.413, 56.166)0.240.397 (−1.043, 1.838)0.59−0.099 (−0.352, 0.154)0.44Alcohol use Prior1.009 (0.636, 1.667)0.970.252 (0.005, 2.483)0.270.941 (0.059, 8.048)0.960.437 (0.002, 10.281)0.580.309 (−1.326, 1.943)0.71−0.064 (−0.343, 0.216)0.66 Current1.097 (0.776, 1.614)0.601.243 (0.263, 4.821)0.760.249 (0.011, 1.891)0.200.373 (0.004, 4.498)0.490.077 (−1.114, 1.269)0.90−0.09 (−0.301, 0.122)0.41Tobacco use Prior1.323 (0.937, 1.863)0.090.86 (0.221, 3.161)0.828.753 (1.696, 97.313)0.0082.079 (0.276, 120.812)0.460.178 (−0.932, 1.288)0.75−0.009 (−0.205, 0.187)0.93 Current1.214 (0.702, 2.143)0.460.203 (0.002, 2.746)0.2712.603 (0.878, 449.477)0.060.565 (0.005, 22.884)0.710.207 (−1.706, 2.12)0.83−0.089 (−0.421, 0.242)0.60Drug use Prior0.918 (0.457, 1.96)0.803.273 (0.084, 67.376)0.440.032 (0, 1.166)0.062.336 (0.01, 195.151)0.65−2.242 (−4.666, 0.181)0.07−0.144 (−0.555, 0.266)0.49 Current0.775 (0.444, 1.398)0.322.04 (0.135, 26.214)0.580.909 (0.047, 10.568)0.941.22 (0.007, 63.605)0.91−1.105 (−2.823, 0.613)0.21−0.069 (−0.377, 0.239)0.66ASA score (per level)3.695 (2.221, 6.593)0.0010.351 (0.004, 3.819)0.473.076 (0.254, 43.645)0.370.85 (0.001, 10.726)0.90−0.031 (−1.975, 1.914)0.98−0.201 (−0.55, 0.149)0.26Number of vertebral levels treated (per level)1.066 (0.922, 1.237)0.410.647 (0.277, 1.184)0.171.242 (0.645, 2.372)0.481.002 (0.219, 9.005)>0.99−0.119 (−0.635, 0.396)0.650.022 (−0.069, 0.112)0.64Vertebral levels involved (vs. thoracic) Lumbar0.804 (0.587, 1.103)0.161.488 (0.416, 5.856)0.540.787 (0.165, 4.005)0.760.608 (0.076, 3.472)0.550.733 (−0.295, 1.761)0.160.089 (−0.095, 0.272)0.35 Sacral1.323 (0.431, 5.954)0.653.695 (0.206, 84.41)0.3633.417 (1.241, 2221.749)0.040.296 (0.001, 23.953)0.65−1.501 (−5.603, 2.602)0.480.14 (−0.598, 0.879)0.71 Thoracolumbar1.203 (0.723, 2.054)0.447.12 (1.028, 73.815)0.0476.515 (0.981, 71.619)0.050.286 (0, 7.385)0.48−0.071 (−1.653, 1.51)0.93−0.082 (−0.366, 0.202)0.57 Lumbosacral1.594 (0.331, 16.314)0.615.795 (0.022, 560.01)0.464.209 (0.009, 1552.407)0.611.381 (0.002, 1412.004)0.910.004 (−6.041, 6.049)>0.990.228 (−0.857, 1.313)0.68 EBL (per mL)1.003 (1.001, 1.007)0.010.999 (0.976, 1.006)0.711.006 (0.999, 1.014)0.101.004 (0.964, 1.018)0.62−0.006 (−0.015, 0.002)0.150 (−0.002, 0.001)0.81 Bilateral approach (vs. unilateral)0.866 (0.54, 1.357)0.530.177 (0.031, 0.863)0.030.547 (0.055, 3.39)0.530.539 (0.04, 4.262)0.520.271 (−1.256, 1.798)0.730.079 (−0.195, 0.354)0.57 Mechanical reduction (vs. balloon)1.868 (1.242, 2.951)0.0030.665 (0.091, 3.363)0.640.24 (0.015, 2.021)0.220.487 (0.017, 4.128)0.530.201 (−1.152, 1.555)0.77−0.16 (−0.402, 0.083)0.20Cement type (vs. confidence) CX01A/B1.429 (0.944, 2.203)0.060.891 (0.202, 3.906)0.880.657 (0.051, 5.94)0.711.06 (0.07, 34.339)0.960.107 (−1.194, 1.408)0.87−0.046 (−0.273, 0.182)0.70 Vertaplex0.652 (0.358, 1.174)0.140.957 (0.073, 10.568)0.977.664 (0.294, 425.195)0.242.994 (0.093, 140.725)0.51−0.701 (−2.616, 1.213)0.47−0.011 (−0.351, 0.329)0.95†, penalized regression used. ‡, ambulatory improvement scale rates 0 as independent ambulation, 1 as assisted ambulation, and 2 as no ambulation (e.g. one unit of improvement represents moving from assisted to independent ambulation). aOR, adjusted odds ratio; ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EBL, estimated blood loss; LOS, length of stay; PSM, propensity score matching; VAS, Visual Analogue Scale. Cohort outcomes before PSM †, penalized regression used. ‡, ambulatory improvement scale rates 0 as independent ambulation, 1 as assisted ambulation, and 2 as no ambulation (e.g. one unit of improvement represents moving from assisted to independent ambulation). aOR, adjusted odds ratio; ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EBL, estimated blood loss; LOS, length of stay; PSM, propensity score matching; VAS, Visual Analogue Scale. Table 3Cohort comparisons pre-PSMVariablesNon-blue zone (n=175)Blue zone (n=15)P value†Age (years)70.00 (63.00, 78.00)75.00 (67.00, 79.00)0.08Sex0.07 Male49 [28]1 [6.7] Female126 [72]14 [93]Service0.20 Neurosurgery64 [37]3 [20] Orthopedics111 [63]12 [80]Race/ethnicity0.80 White122 [70]11 [73] Black6 [3.4]0 [0] Hispanic35 [20]0 [0] Asian10 [5.7]3 [20] Native2 [1.1]1 [6.7]BMI (kg/m2)27.00 (23.70, 31.39)24.48 (22.50, 30.62)0.50Diabetes0.90 No143 [82]12 [80] Yes32 [18]3 [20]Osteoporosis0.14 No41 [23]1 [6.7] Yes134 [77]14 [93]COPD>0.90 No150 [86]13 [87] Yes25 [14]2 [13]CHF0.044 No151 [86]10 [67] Yes24 [14]5 [33]Hypertension0.13 No49 [28]7 [47] Yes126 [72]8 [53]CKD (grade)0.30 None149 [85]11 [73] 12 [1.1]0 [0] 24 [2.3]1 [6.7] 314 [8.0]3 [20] 43 [1.7]0 [0] 53 [1.7]0 [0] Non-spine malignancy0.14 None123 [70]8 [53] Current22 [13]2 [13] Previous30 [17]5 [33]Alcohol use0.008 None116 [66]15 [100] Prior19 [11]0 [0] Current40 [23]0 [0]Tobacco use0.006 None92 [53]14 [93] Prior65 [37]0 [0] Current18 [10]1 [6.7]Drug use0.13 None151 [86]15 [100] Prior9 [5.1]0 [0] Current15 [8.6]0 [0]ASA score>0.90 246 [26]4 [27] 3122 [70]10 [67] 47 [4.0]1 [6.7]Vertebral levels involved0.20 Thoracic64 [37]4 [27] Lumbar80 [46]6 [40] Sacral1 [0.6]1 [6.7] Thoracolumbar29 [17]4 [27] Lumbosacral1 [0.6]0 [0]EBL (mL)5.00 (1.00, 10.00)5.00 (5.00, 50.00)0.30Surgical approach0.90 Unilateral27 [16]2 [13] Bilateral147 [84]13 [87]Reduction technique>0.90 Balloon129 [74]11 [73] Mechanical45 [26]4 [27]Cement type0.3 Confidence (DePuy)97 [57]6 [43] CX01A/B (Medtronic)55 [32]6 [43] Vertaplex (Stryker)18 [11]2 [14]DEXA t-score−1.50 (−2.70, −0.30)−1.60 (−2.20, −1.30)0.7Data are presented as median (Q1, Q3) or n [%]. †, Wilcoxon rank sum test or Welch two-sample t-test. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; PSM, propensity score matching. Cohort comparisons pre-PSM Data are presented as median (Q1, Q3) or n [%]. †, Wilcoxon rank sum test or Welch two-sample t-test. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; PSM, propensity score matching. Unmatched blue zone and non-blue zone cohorts In the unmatched cohort, there were 15 blue zone patients and 175 non-blue zone patients. Pre-PSM, the blue zone mean cohort age was 73.4±7.95 years and the non-blue zone mean cohort age was 69.3±12.6 years. Regarding BMI, the blue zone mean BMI was 26.5±5.67 kg/m2 and the non-blue zone mean BMI was 27.6±5.34 kg/m2. Sex distribution and racial composition did not differ significantly. Mean LOS was 2.67±2.05 days for blue zone patients and 1.07±1.99 days for non-blue zone patients. Blue zone patients showcased a mean post-operative improvement in pain score of 2.23±2.5 and a mean post-operative improvement in ambulatory function of 0.43±0.5. For non-blue zone patients, the mean post-operative improvement in pain score was 2.23±2.8 and the mean post-operative improvement in ambulatory function was 0.21±0.4. Among blue zone patients, 66.7% (n=10) identified as SDA. Blue zone patients had lower rates of alcohol use (0% vs. 34%, P=0.03) and tobacco use (7% vs. 47%, P<0.01). No other significant findings between blue zone and non-blue zone patients were appreciated pre-PSM regarding demographics, comorbidities, and surgical outcomes (Table 2). Matched blue zone and non-blue zone cohorts After 1:3 PSM, there were 14 blue zone patients and 42 non-blue zone patients. The blue zone cohort’s mean age was 74.7±6.34 and the non-blue zone cohort’s mean age was 68.6±6.34 years. Regarding BMI, the blue zone mean BMI was 26.6±5.8 and the non-blue zone mean BMI was 27.8±5.8 kg/m2. Sex distribution and racial composition did not differ significantly. Mean LOS was 2.71±4.4 days for blue zone patients and 0.71±1.5 days for non-blue zone patients. Blue zone patients showcased a mean post-operative improvement in pain score of 1.75±1.9 and a mean post-operative improvement in ambulatory function of 0.38±0.5. For non-blue zone patients, the mean post-operative improvement in pain score was 2.27±2.4 and the mean post-operative improvement in ambulatory function was 0.1±0.4. Consistent with the unmatched analysis, post-PSM blue zone patients demonstrated lower rates of alcohol use (0% vs. 64%, P<0.01) and tobacco use (7% vs. 43%, P=0.04). All other variables were similarly distributed and outcomes such as LOS, surgical complications, readmission, reoperation, ambulatory and pain improvement, and mortality were not significantly different between groups. Post-matching cohort outcomes are summarized in Table 4. Table 5 showcases cohort comparisons post-PSM. Figures 1,2 showcases 10-year Kaplan-Meier survival curves for non-blue zone and blue zone patients both pre- and post-PSM. Table 4Cohort outcomes after PSMVariablesLOSVAS pain score improvementAmbulatory improvement†Mean ratio (95% CI)P valueβ (95% CI)P valueβ (95% CI)P valueBlue zone (vs. non-blue zone)2.88 (1.036, 8.007)0.056−0.708 (−4.029, 2.612)0.680.029 (−0.633, 0.692)0.93Age (per year)1.023 (0.991, 1.055)0.170.024 (−0.076, 0.123)0.64−0.002 (−0.022, 0.018)0.87Female (vs. male)1.561 (0.736, 3.311)0.262.028 (−0.394, 4.449)0.120.116 (−0.372, 0.603)0.65Orthopedic service (vs. neurosurgery)0.895 (0.394, 2.034)0.79−3.332 (−6.121, −0.542)0.030.098 (−0.435, 0.63)0.72Race Black (vs. White)0.987 (0.026, 37.358)0.991.222 (−7.894, 10.338)0.80−0.281 (−2.636, 2.073)0.82 Hispanic (vs. White)1.155 (0.521, 2.561)0.730.518 (−2.276, 3.312)0.720.027 (−0.489, 0.543)0.92 Asian (vs. White)0.611 (0.127, 2.934)0.55−0.749 (−5.916, 4.418)0.780.561 (−0.456, 1.578)0.29 Native (vs. White)0.811 (0.118, 5.565)0.83−1.789 (−8.396, 4.818)0.60−0.298 (−1.546, 0.95)0.65BMI (per kg/m2)0.992 (0.946, 1.041)0.760.047 (−0.14, 0.234)0.63−0.011 (−0.042, 0.02)0.48Diabetes1.284 (0.385, 4.28)0.693.118 (−0.899, 7.135)0.140.324 (−0.456, 1.104)0.43Osteoporosis0.39 (0.177, 0.859)0.03−0.603 (−3.132, 1.927)0.65−0.067 (−0.578, 0.444)0.80COPD0.077 (0.011, 0.53)0.02−2.376 (−8.709, 3.956)0.47−0.422 (−1.671, 0.827)0.52CHF1.224 (0.476, 3.145)0.680.897 (−2.128, 3.921)0.57−0.042 (−0.654, 0.569)0.89Hypertension0.791 (0.296, 2.111)0.64−1.953 (−5.134, 1.228)0.24−0.454 (−1.09, 0.183)0.18CKD3.93 (0.308, 50.19)0.300.272 (−7.369, 7.913)0.95−0.361 (−2.011, 1.29)0.67Non spine malignancy Prior0.82 (0.334, 2.013)0.671.054 (−1.833, 3.941)0.480.033 (−0.549, 0.615)0.91 Current0.723 (0.299, 1.747)0.481.528 (−1.393, 4.45)0.32−0.183 (−0.755, 0.389)0.54Tobacco use Prior1.316 (0.626, 2.769)0.48−0.325 (−2.872, 2.221)0.810.15 (−0.332, 0.632)0.55 Current6.273 (0.676, 58.237)0.12––––Drug use Prior0.793 (0.214, 2.931)0.731.075 (−3.68, 5.831)0.66−0.09 (−0.938, 0.757)0.84 Current0.753 (0.31, 1.831)0.54−1.434 (−4.33, 1.462)0.340.068 (−0.508, 0.644)0.82ASA score (per level)1.222 (0.428, 3.493)0.710.24 (−1.442, 1.922)0.780.152 (−0.188, 0.492)0.39Number of vertebral levels treated (per level)1.434 (1.115, 1.844)0.01−0.136 (−0.946, 0.674)0.750.105 (−0.058, 0.268)0.22Vertebral levels involved (vs. thoracic) Lumbar1.463 (0.703, 3.043)0.320.333 (−2.118, 2.785)0.790.233 (−0.241, 0.708)0.35 Sacral1.161 (0.25, 5.4)0.85−2.444 (−7.527, 2.64)0.360.129 (−0.867, 1.125)0.80 Thoracolumbar0.701 (0.253, 1.94)0.500.251 (−3.048, 3.549)0.88−0.154 (−0.813, 0.506)0.65EBL (per mL)1.006 (0.99, 1.022)0.47−0.027 (−0.078, 0.024)0.320 (−0.01, 0.01)>0.99Bilateral approach (vs. unilateral)0.784 (0.104, 5.881)0.821.2 (−5.431, 7.831)0.73−0.053 (−1.359, 1.253)0.94Mechanical reduction (vs. balloon)1.974 (0.417, 9.342)0.402.003 (−3.312, 7.317)0.47−0.661 (−1.668, 0.346)0.21Cement type (vs. confidence) CX01A/B0.73 (0.189, 2.817)0.65−4.442 (−8.79, −0.094)0.059−0.54 (−1.415, 0.335)0.24 Vertaplex0.521 (0.09, 3.018)0.48−4.478 (−10.312, 1.356)0.150.478 (−0.66, 1.617)0.42†, ambulatory improvement scale rates 0 as independent ambulation, 1 as assisted ambulation, and 2 as no ambulation (e.g. one unit of improvement represents moving from assisted to independent ambulation). ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EBL, estimated blood loss; LOS, length of stay; PSM, propensity score matching; VAS, Visual Analogue Scale. Cohort outcomes after PSM †, ambulatory improvement scale rates 0 as independent ambulation, 1 as assisted ambulation, and 2 as no ambulation (e.g. one unit of improvement represents moving from assisted to independent ambulation). ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CI, confidence interval; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; EBL, estimated blood loss; LOS, length of stay; PSM, propensity score matching; VAS, Visual Analogue Scale. Table 5Cohort comparisons post-PSMVariablesNon-blue zone (n=45)Blue zone (n=15)P value†Age (years)72.00 (66.00, 81.00)75.00 (67.00, 79.00)0.40Sex0.14 Male11 [24]1 [6.7] Female34 [76]14 [93]Service0.30 Neurosurgery15 [33]3 [20] Orthopedics30 [67]12 [80]Race/ethnicity0.20 White39 [87]11 [73] Black0 [0]0 [0] Hispanic0 [0]0 [0] Asian6 [13]3 [20] Native0 [0]1 [6.7]BMI (kg/m2)27.00 (22.90, 31.10)24.48 (22.50, 30.62)>0.90Diabetes>0.90 No36 [80]12 [80] Yes9 [20]3 [20]Osteoporosis0.14 No11 [24]1 [6.7] Yes34 [76]14 [93]COPD0.80 No40 [89]13 [87] Yes5 [11]2 [13]CHF0.60 No33 [73]10 [67] Yes12 [27]5 [33]Hypertension0.30 No14 [31]7 [47] Yes31 [69]8 [53]CKD (grade)0.04 None42 [93]11 [73] 10 [0]0 [0] 21 [2.2]1 [6.7] 32 [4.4]3 [20] 40 [0]0 [0] 50 [0]0 [0]Non spine malignancy0.20 None32 [71]8 [53] Current4 [8.9]2 [13] Previous9 [20]5 [33]Alcohol use– None45 [100]15 [100] Prior0 [0]0 [0] Current0 [0]0 [0]Tobacco use0.80 None41 [91]14 [93] Prior0 [0]0 [0] Current4 [8.9]1 [6.7]Drug use– None45 [100]15 [100] Prior0 [0]0 [0] Current0 [0]0 [0]ASA score0.90 211 [24]4 [27] 331 [69]10 [67] 43 [6.7]1 [6.7]Vertebral levels involved0.10 Thoracic23 [51]4 [27] Lumbar14 [31]6 [40] Sacral0 [0]1 [6.7] Thoracolumbar8 [18]4 [27] Lumbosacral0 [0]0 [0]EBL (mL)1.00 (1.00, 5.00)5.00 (5.00, 50.00)0.10Surgical approach0.40 Unilateral11 [24]2 [13] Bilateral34 [76]13 [87]Reduction technique0.90 Balloon32 [71]11 [73] Mechanical13 [29]4 [27]Cement type0.20 Confidence (DePuy)26 [60]6 [43] CX01A/B (Medtronic)14 [33]6 [43] Vertaplex (Stryker)3 [7.0]2 [14]DEXA t-score−1.60 (−3.20, −0.20)−1.60 (−2.20, −1.30)0.50Data are presented as median (Q1, Q3) or n [%]. †, Wilcoxon rank sum test or Welch two-sample t-test. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; PSM, propensity score matching. Cohort comparisons post-PSM Data are presented as median (Q1, Q3) or n [%]. †, Wilcoxon rank sum test or Welch two-sample t-test. ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; DEXA, dual-energy X-ray absorptiometry; EBL, estimated blood loss; PSM, propensity score matching. Ten-year Kaplan-Meier survival curve for non-blue zone and blue zone patients pre propensity score match. Ten-year Kaplan-Meier survival curve for non-blue zone and blue zone patients post propensity score match. Discussion Demographics, health metrics, and chronic medical conditions This study investigated how patient demographics, health status, surgical factors, and geographic location, particularly residence in a blue zone, affected outcomes following kyphoplasty for VCFs. Management for VCFs has frequently involved non-surgical modalities including medication and physical therapy, attempting to manage related complications of osteoporosis and age that often contributed to the initial VCF (7,8). However, kyphoplasty procedures have become widely utilized for treating refractory symptomatic VCFs, particularly for pain management and preventing progression of further vertebral deformity or compression (9). Literature has established potential relationships regarding comorbidities such as osteoporosis, diabetes, smoking status, and obesity with post-operative improvement and symptomatic management (7,10,11). The present findings contribute to the ongoing conversation regarding the impact of modifiable lifestyle interventions on kyphoplasty outcomes, demonstrating a significant relationship between the presence of tobacco and increased 30-day readmission rates, as well as a significant relationship between diabetes and an increased LOS. Additionally, Cabrera et al. found that age, ASA class 2, and partially dependent functional status were associated with increased 30 day-readmission following both vertebroplasty and kyphoplasty procedures (12). This study similarly demonstrated that age and ASA score were both associated with a significant increase in LOS. Collectively, this information reinforces the significance of traditional risk factors including age, ASA score, diabetes, and tobacco use as key indicators of kyphoplasty outcomes. While prior research has compared biomechanics and clinical outcomes of cement compositions, such as polymethylmethacrylate (PMMA), calcium phosphate, or glass polyalkenoate cements, no studies have evaluated the effect of the specific manufacture/type on kyphoplasty outcomes (13,14). Our study analyzed three commonly used products: CX01A/B Xpede Bone Cement (Medtronic, Memphis, TN, USA), Confidence Spinal Cement System (DePuy Synthes, Raynham, MA, USA), and VertaPlex Cement (Stryker, Kalamazoo, MI, USA). All were PMMA-based, allowing for cohesive comparison. Comparisons between manufacturers of neurosurgical procedural tools have shown effects on clinical outcomes; however, our study found no significant differences between cement manufacturers and kyphoplasty outcomes (15). Such findings suggest that differing PMMA-based cement selection remains largely a matter of cement preference or procedure type. Within 30 days postoperatively, 21% of our cohort developed a post-operative surgical complication, including neurologic deficits, vertebral fractures, and infection. While 30-day complication rates are variable between studies, they have been reported to be as high as 16.2%, contrasting our cohort’s rate of complications (16). Notably, nearly 10% of patients specifically developed a new neurologic deficit (paralysis, altered mental status, and paresthesia), constituting nearly half (47.5%) of all post-op complications. This relatively high proportion may reflect variable presentations of transient postoperative delirium commonly seen in elderly patients with comorbidities, consistent with our population, as opposed to more severe, debilitating neurologic impairment. Furthermore, despite the clinical relevance of vertebral level and degree of height loss for preoperative planning, our study found no significant associations between these factors and clinical outcomes, consistent with prior reports of comparable kyphoplasty effectiveness across VCF locations and anterior vertebral compression ratio (17,18). However, thoracolumbar involvement was significantly associated with increased new neurological complications, suggesting this region may warrant particular caution during surgical planning. In terms of readmission, our cohort demonstrated a 30-day readmission rate of 11%, aligning with published rates of around 10.8% (19). At long-term follow-up, 21 patients were deceased. These deaths were not directly attributable to kyphoplasty, supporting prior evidence of low procedure-related mortality (19). Nonetheless, as VCF patients are often older with comorbidities, kyphoplasty investigations have primarily focused on outcomes within 2–3 years postoperatively (20). Long-term survival and functional trajectories after kyphoplasty remain poorly understood and warrant further research. Overall, the findings of this study highlight both the benefits and risks of kyphoplasty. While generally safe, the procedure carries a notable risk of complications, particularly neurologic events, and underscores the importance of thorough preoperative counseling and risk stratification when considering surgical management for VCFs. Blue zone factors Prior research has highlighted the influence of blue zones aspects on longevity, largely attributed to the “Power of 9” lifestyle factors, but there has been limited exploration of these variables on surgical outcomes (6). Ruckle et al. found that blue zone patients experienced delayed onset of frailty and osteoporotic hip fractures by 12 years, as well as lower rates of postoperative mortality compared to non-blue zone patients (21). In our pre-PSM match cohort, blue zone patients undergoing kyphoplasty were significantly less likely to utilize alcohol and tobacco, and fewer in number, consistent with the protective health effects and delayed onset of chronic disease observed in blue zones. These findings support the presence of healthier lifestyle habits at baseline in the blue zone cohort with respect to lower alcohol and tobacco use. However, post-PSM, no significant differences in kyphoplasty outcomes were identified. Although blue zone patients may experience later onset of frailty and osteoporotic VCFs, our results showcase that despite more favorable lifestyle habits, no detectable association was identified between blue zone status and kyphoplasty outcomes (22). A potential explanation is that even after post-PSM, blue zone patients undergoing kyphoplasty may represent a highly selective subset, likely those with the most severe or refractory cases of VCFs, limiting generalizability and may not reflect the true protective influence of the blue zone lifestyle on spine surgery. When examining blue zones, it is essential to acknowledge that each location possesses unique cultural and demographic characteristics. Our analysis focused on Loma Linda, one of the five original blue zones, which is distinct in its large population of SDAs (23). SDA practices, such as high rates of vegetarianism and abstinence from alcohol and tobacco, may act as potential confounders when evaluating health outcomes. Although we incorporated religion (including SDA affiliation) as a variable in our analysis, no significant associations were found. However, SDA affiliation may overlap substantially with blue zone classification itself, and inclusion of religion as a covariate may introduce multicollinearity or overadjustment. Subsequently, this factor should be considered when generalizing the present findings beyond Loma Linda. Additionally, despite the validation of long-lived individuals in blue zones, such as centenarians, broader concerns regarding the validity and applicability of the blue zone concept have also been raised (24). Newman et al. highlighted potential confounding influences regarding blue zone validity and clinical applicability, citing factors of differing geography, socioeconomic status, literacy, crime rates, and even inconsistencies in reported birth, age, and death records (25). Our study faced important constraints, as the duration of patients’ residence in blue zones was not available. As recent relocation into the region could potentially bias classification of blue zone exposure, extended chart review was additionally performed to evaluate longitudinal residence patterns within the medical record. However, heterogeneity of the cohort and significant exposure misclassification remains possible and may have attenuated the ability to detect true associations. Additionally, lifestyle-specific variables such as detailed dietary patterns, nutritional intake, vitamin supplementation, and levels of physical activity were not analyzed. Consequently, reliance on geographic residence as a proxy for blue zone lifestyle may be a critical oversimplification without further validation that may undermine the primary research question and weaken downstream interpretations. While our sample size precluded a mediation analysis, future studies with larger cohorts may benefit from such an approach to clarify the true effect of blue zones on kyphoplasty. However, we hope our study contributes to building a more rigorous and evidence-based understanding of the relationship between blue zones and surgical outcomes. Limitations Beyond the intrinsic challenges of evaluating blue zone variables, this study is subject to several important limitations. Firstly, our patient population was obtained from a single institution. While our racial and ethnic distribution reflects the demographics of our institution, it is not necessarily representative of larger demographics across the United States, potentially limiting the generalizability of this study. For example, despite the lack of detectable associations regarding race, only 3% of our cohort was comprised of Black patients. Literature has demonstrated significant disparities in prevalence and access to kyphoplasty in Medicare/Medicaid dual-eligible and Black patients, and such considerations remain essential when determining clinical symptom severity or surgical inadequacy (10,26). Furthermore, the overall sample size of our patient population was relatively small, particularly the blue zone group, which may have reduced statistical power and constrained the scope of PSM and subgroup analysis. As such, the lack of significant differences between both groups is an exploratory finding which may reflect limited power rather than a lack of significant association, restricting the interpretation of an absence of statistically significant findings as evidence of no association. Additionally, although the older age of our cohort is expected given the established relationship between age and VCFs, the skewed demographic age makeup of our population remains a significant limitation in terms of applicability of our results to more diverse populations. Nevertheless, our study remains among the most comprehensive to date, including unique variables not previously investigated. Furthermore, some patients with severe osteopenia may not have been formally diagnosed with osteoporosis, which could introduce misclassification bias. Finally, despite the inherently subjective nature of pain, patient pain scores were self-reported, introducing potential variability and bias. Such outcome measures may have been influenced not only by VCF severity, but also by individual differences in pain tolerance, analgesic usage, and psychosocial factors such as anxiety or mental health status. Conclusions This study evaluated the impact of patient demographics, surgical factors, and blue zone residence on outcomes following kyphoplasty for VCFs. Traditional risk factors including age, ASA score, diabetes, and tobacco remain the primary indicators of kyphoplasty outcomes. While blue zone patients demonstrated lower alcohol and tobacco use, post-PSM demonstrated a lack of detectable association in perioperative outcomes, functional status and pain scores, complications, or long-term survival compared to non-blue zone patients. These results suggest that while lifestyle and environmental factors may delay onset of chronic disease and VCF occurrence, no detectable association was identified between blue zone residence and kyphoplasty outcomes. The interpretation of these findings should be done cautiously given the small sample size of the blue zone cohort and the use of geographic residence as a proxy for blue zone lifestyle habits. Future research utilizing larger, more diverse cohorts and detailed lifestyle assessments are needed to better clarify the effect of social determinants of health and blue zone lifestyle on kyphoplasty outcomes. Supplementary The article’s supplementary files as Available at https://jss.amegroups.com/article/view/10.21037/jss-2026-0095/prf