Spring-mediated cranioplasty (SMC) is a procedure used to treat nonsyndromic craniosynostosis, most commonly sagittal synostosis.1 Our institution preferentially employs SMC for surgical treatment of infantile nonsyndromic sagittal craniosynostosis because of some benefits over open and endoscopic techniques.2–5 Perioperative outcomes are favorable for SMC compared with other surgical approaches, such as cranial vault remodeling3,6,7 and strip craniectomy with barrel staving,5 with SMC demonstrating consistent postoperative outcomes and lower estimated blood loss (EBL), procedure duration, and hospital length of stay (LOS).3
Despite overall consistency, there can be variability of perioperative outcomes and complications. Furthermore, although blood loss is generally low and the need for transfusion infrequent, factors implicated in these outcomes are not clearly established. To our knowledge, parietal bone, or calvarial thickness of any kind, has not been studied alongside perioperative outcomes in patients undergoing procedures involving craniectomy of a closed suture. Therefore, we have investigated the impact of parietal bone thickness on changes in the cephalic index in patients who underwent SMC, with particular focus on the relationship between spring parameters and parietal bone thickness. In exploratory analyses, we identified relationships between parietal bone thickness and perioperative outcomes, which have formed the basis of the present study. The purpose of this study was to characterize relationships between parietal bone thickness and perioperative outcomes, including the need for intraoperative transfusion, EBL, hospital LOS, and complications, in patients who underwent SMC for nonsyndromic sagittal craniosynostosis.
Methods
Patients
This retrospective study included patients with nonsyndromic sagittal craniosynostosis who underwent SMC at our institution between 2011 and 2021. Exclusion criteria were syndromic craniosynostosis or a history of prior cranial vault remodeling. This study was approved by the IRB at the Children’s Hospital of Philadelphia.
Operative Technique
The operative technique for SMC has been previously described.1,2,8 In brief, the cranium is exposed along the length of the sagittal suture and an approximately 1.5-cm-wide suturectomy is performed with an ultrasonic bone cutter by the neurosurgeon. Two or three springs are selected for length, width, and spring forces by the attending plastic surgeon, and osteotomy notches are created along the strip craniectomy margins to engage the footplates of each spring. Springs are placed 1 cm anterior to the lambdoid sutures and 1 cm posterior to the coronal sutures. A third spring is placed at the midpoint of the parietal bone if placement does not lead to prominence in the sagittal plane, spring crowding, or increased risk of spring dislodgment. The resulting craniectomy strip is morcellized and placed over the sagittal bone suture, and deep to the springs, to facilitate osteogenesis beneath cranial springs.
Clinical Variables
Clinical data were obtained from patient medical records, including demographic information (age at surgery, sex, and race), spring forces, and the number of springs placed. Perioperative variables, including the EBL, need for intraoperative blood transfusion with packed red blood cells (PRBCs), procedure duration, hospital LOS, and postoperative complications, were also evaluated. General criteria for transfusion included hemodynamic instability, decreased hemoglobin (typically < 7 g/dL), or significant EBL (typically > 50 mL). Postoperative complications included hardware/spring failure, cephalohematoma, pseudomeningocele, seroma, and CSF leakage.
Image Analysis
Parietal bone thickness was determined on patient preoperative CT using Materialise Mimics version 23 (Materialise). Images were appropriately thresholded using established methods for evaluating bone in Materialise Mimics software. Parietal bones were isolated from the remainder of the cranium through manual imaging segmentation. Three-dimensional parietal bone renderings of each patient were imported into Materialise 3-matic version 15 (Materialise).
Thickness analysis was performed through an automated thickness analysis function in Materialise 3-matic to determine the mean, median, and interquartile range for the thickness of each parietal bone.9 Parietal bone thickness heat maps were generated for each patient as a quality control measure to visualize areas of bone inadvertently registered as missing or overestimated in thickness. Thickness analysis in Materialise 3-matic is demonstrated here: https://www.materialise.com/en/academy-medical/mimics-innovation-suite/tutorials/analyze-wall-thickness.
Parietal bone thickness was determined at 27 points in relation to the suture (Fig. 1). Anterior parietal, midparietal, and posterior parietal were defined as 1.0 cm posterior to the coronal suture, the middle of the parietal bone, and 1.0 cm anterior to the lambdoid suture, respectively, corresponding to the typical spring placement locations at our institution. The "measure" tool in Materialise 3-matics was used to determine distances mapped to each 3D parietal bone rendering. Using the same tool, points were marked at the anterior parietal, midparietal, and posterior parietal at distances of 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm from the suture bilaterally. The thickness at specific points was determined using the “analyze locally” tool in Materialise 3-matics. Results presented at suture-related distances (0.5 cm) represent averages from the anterior parietal, midparietal, and posterior parietal measurements bilaterally.
Parietal bone thickness renderings for lateral (A) and overhead (B) views of 3D rendering (left); heat map of thickness analysis (center); and heat map with thickness points of interest anteriorly, medially, and posteriorly at the suture and 0.5, 1.0, 1.5, and 2.0 cm from the suture (right). Anterior parietal was defined as 1 cm posterior to the coronal sutures, midparietal as the middle of the parietal bone, and posterior parietal as 1 cm anterior to the lambdoid suture, corresponding to the ideal placement of spring expansion devices. Images generated in Materialise 3-matic version 15 (Materialise). Figure is available in color online only.
Statistical Analysis
Demographic data were analyzed with descriptive statistics. Parietal bone thickness measurements were compared using Wilcoxon rank-sum tests. EBL and EBL per kilogram were compared with parietal bone thickness via Spearman’s correlations and multivariate linear regression models. The parietal bone thicknesses of the transfusion and nontransfusion groups were compared using the Mann-Whitney U-test. The need for intraoperative transfusion was predicted by parietal bone thickness with multivariate logistic regression models. Statistical analysis was performed in JASP 0.15 for Windows (https://jasp-stats.org).10
Results
Patient Demographics
During the study interval, 124 patients presenting with nonsyndromic sagittal craniosynostosis who underwent SMC at the Children’s Hospital of Philadelphia were included (Table 1). Patients underwent surgery at a mean age ± SD of 3.59 ± 0.87 months. The majority of patients were male (n = 94, 75.8%) and White (n = 98, 79.0%).
Patient demographics
| Value | |
|---|---|
| No. of pts | 124 |
| Mean age ± SD, mos | 3.59 ± 0.87 |
| Mean weight ± SD, kg | 6.34 ± 1.01 |
| Sex, n (%) | |
| M | 94 (75.8) |
| F | 30 (24.2) |
| Race, n (%) | |
| White | 98 (79.0) |
| Black/African American | 8 (6.5) |
| Hispanic/Latino | 7 (5.6) |
| Other | 7 (5.6) |
| Unknown | 2 (1.6) |
Pts = patients.
Clinical Outcomes
Perioperative outcomes and complications are detailed in Table 2. Patients had mean amounts of intraoperative EBL of 49.5 ± 43.93 mL and weight-corrected EBL per kilogram of 7.21 ± 7.01 mL/kg. Twenty-nine patients (23.4%) received an intraoperative transfusion of a mean of 145.27 ± 82.09 mL of blood per transfusion and mean weight-corrected transfusion of 22.69 ± 12.15 mL/kg. Sixty patients (48.4%) received tranexamic acid. The mean operative procedure duration was 1.73 ± 0.79 hours, and patients remained hospitalized for a mean of 1.72 ± 0.88 days. Ten patients (8.1%) experienced complications, including 2 patients with dislodged springs, 2 patients with cephalohematoma, 2 patients with pseudomeningocele, 2 patients with seroma, and 2 patients with a CSF leak. Nine patients (7.3%) underwent reoperation (beyond the planned spring removal) for head shape relapse (n = 5), pseudomeningocele repair (n = 2), or removal/revision of a dislodged spring (n = 2).
Perioperative outcomes and complications
| Value | |
|---|---|
| Mean EBL, mL | 49.47 ± 43.93 |
| Mean EBL/kg, mL/kg | 7.21 ± 7.01 |
| Transfusion, n (%) | 29 (23.4) |
| Mean amount transfused, mL | 145.27 ± 82.09 |
| Mean amount transfused/kg, mL/kg | 22.69 ± 12.15 |
| Mean op duration, hrs | 1.73 ± 0.79 |
| Mean LOS, days | 1.72 ± 0.88 |
| Complications, n (%) | 10 (8.1) |
| Hardware failure (dislodged spring) | 2 |
| Hematoma | 2 |
| Pseudomeningocele | 2 |
| Seroma | 2 |
| CSF leak | 2 |
| Reop, n (%) | 9 (7.3) |
| Relapse/head shape concern | 5 |
| Pseudomeningocele repair | 2 |
| Removal of dislodged spring | 2 |
Mean values are presented as the mean ± SD.
Parietal Bone Thickness
Overall, 98 patients in the cohort had obtainable preoperative CT imaging for parietal bone thickness analysis (Table 3). Isolated parietal bones were a mean of 1.83 ± 0.38 mm thick, and the mean maximum thickness was 3.85 ± 0.74 mm.
Parietal bone thickness at suture-specific points
| Mean ± SD | ||||
|---|---|---|---|---|
| Anterior | Middle | Posterior | Overall | |
| Suture line | 1.608 ± 0.488 | 3.032 ± 0.903 | 2.534 ± 0.902 | 2.363 ± 0.584*** |
| 0.5 cm rt | 1.770 ± 0.386 | 2.239 ± 0.648 | 2.300 ± 0.530 | |
| 0.5 cm lt | 1.822 ± 0.379 | 2.279 ± 0.638 | 2.247 ± 0.525 | |
| 0.5-cm average | 1.796 ± 0.383 | 2.259 ± 0.643 | 2.274 ± 0.528 | 2.108 ± 0.349*** |
| 1.0 cm rt | 1.685 ± 0.376 | 1.669 ± 0.651 | 2.078 ± 0.654 | |
| 1.0 cm lt | 1.668 ± 0.437 | 1.858 ± 0.494 | 2.106 ± 0.583 | |
| 1-cm average | 1.677 ± 0.407 | 1.764 ± 0.573 | 2.092 ± 0.619 | 1.840 ± 0.376*** |
| 1.5 cm rt | 1.745 ± 0.340 | 1.831 ± 0.549 | 2.189 ± 0.605 | |
| 1.5 cm lt | 1.735 ± 0.407 | 1.850 ± 0.551 | 2.165 ± 0.539 | |
| 1.5-cm average | 1.740 ± 0.374 | 1.841 ± 0.550 | 2.177 ± 0.572 | 1.914 ± 0.340** |
| 2.0 cm rt | 1.780 ± 0.401 | 1.837 ± 0.452 | 2.241 ± 0.502 | |
| 2.0 cm lt | 1.795 ± 0.508 | 1.945 ± 0.615 | 2.126 ± 0.556 | |
| 2.0-cm average | 1.788 ± 455 | 1.891 ± 0.534 | 2.183 ± 0.529 | 1.945 ± 0.352 |
| Overall | 1.734 ± 0.298*** | 2.063 ± 0.421** | 2.221 ± 0.443** | |
The measurements 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm refer to distances from the suture. Statistical significance between group means with the Wilcoxon rank-sum test are denoted as
p < 0.01,
p < 0.001.
Parietal bone thickness varied topographically based on anteroposterior positioning and distance from the suture. The anterior parietal (1 cm posterior to the coronal sutures) was significantly thinner than the midparietal (1.788 ± 0.455 mm vs 1.891 ± 0.534 mm, p < 0.001), and the posterior parietal (1 cm anterior to the lambdoid sutures) was significantly thicker than the midparietal (2.183 ± 0.529 mm vs 1.891 ± 0.534 mm, p = 0.002).
Parietal bone was thickest at the suture line (2.363 ± 0.584 mm), and thickness decreased from the suture to 0.5 cm from the suture (2.108 ± 0.349 mm) and 1.0 cm from the suture (1.840 ± 0.376 mm) (all p < 0.001). Thickness increased from 1.0 cm to 1.5 cm from the suture line (1.914 ± 0.340 mm, p < 0.001) and remained comparable with 1.5 cm to 2.0 cm from the suture line (1.945 ± 0.352 mm, p = 0.258) (Table 3). Age was not associated with the mean parietal bone thickness (ρ = −0.079, p = 0.438), maximum parietal bone thickness (ρ = −0.036, p = 0.757), or any suture-related thickness measurements (0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm, all p > 0.05).
Parietal Bone Thickness and EBL
Parietal bone thickness was associated with EBL, as detailed in Table 4. EBL and parietal bone thickness were significantly associated at 0.5 cm (ρ = 0.376, p < 0.001), 1.0 cm (ρ = 0.324, p = 0.007), and 1.5 cm (ρ = 0.272, p = 0.024) from the suture line, but not at the suture line itself (ρ = 0.186, p = 0.125) or 2.0 cm from the suture line (ρ = 0.220, p = 0.070). EBL was also associated with the mean parietal bone thickness (ρ = 0.316, p = 0.003) and maximum parietal bone thickness (p = 0.326, p = 0.006) but not age (ρ = −0.106, p = 0.265). When evaluating the associations of EBL and controlling for weight (mL/kg), similar associations were observed, with thickness at 0.5 cm (ρ = 0.331, p = 0.004) and 1.0 cm (ρ = 0.245, p = 0.033) from the suture being the most significant.
Independent associations of blood loss and parietal bone thickness
| EBL (ρ) | p Value | EBL/kg (ρ) | p Value | |
|---|---|---|---|---|
| Suture line | 0.186 | 0.125 | 0.194 | 0.094 |
| 0.5 cm | 0.376 | <0.001*** | 0.331 | 0.004** |
| 1.0 cm | 0.324 | 0.007** | 0.245 | 0.033* |
| 1.5 cm | 0.272 | 0.024* | 0.191 | 0.098 |
| 2.0 cm | 0.22 | 0.070 | 0.147 | 0.205 |
| Mean | 0.316 | 0.003** | 0.181 | 0.074 |
| Maximum | 0.326 | 0.006** | 0.214 | 0.061 |
| Age | −0.106 | 0.265 | −0.243 | 0.007** |
The measurements 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm refer to distances from the suture. Statistically significant correlations between EBL or EBL/kg and parietal bone thickness via Spearman’s correlations (ρ) are denoted as
p < 0.05
p < 0.01,
p < 0.001.
Multivariate linear regression models controlling for age, sex, and race revealed that EBL was associated with parietal bone thickness at 0.5 cm from the suture line, with an effect size of 50.20 (95% CI 20.03–80.37) and overall model fit of R2 = 0.154 (p < 0.001) (Table 5). EBL was also significantly associated with thickness 1.0 cm (p = 0.014) and 1.5 cm (p = 0.039) from the suture line, but not at the suture line (p = 0.285) or 2.0 cm from the suture line (p = 0.397). The maximum parietal bone thickness was also predictive of EBL (p = 0.019). The mean parietal bone thickness on multivariate linear regression models, as well as age on univariate linear regression models, was not associated with EBL in this cohort (both p > 0.05).
Multivariate linear regression models of EBL and parietal bone thickness
| EBL/kg | EBL | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| β | 95% CI | R2 | F | p Value | β | 95% CI | R2 | F | p Value | |
| Suture line | 1.429 | −1.678 to 4.536 | 0.066 | 1.262 | 0.362 | 10.719 | −9.127 to 30.565 | 0.026 | 0.424 | 0.285 |
| 0.5 cm | 7.482 | 2.847 to 12.117 | 0.176 | 3.782 | 0.002** | 50.195 | 20.025 to 80.365 | 0.154 | 2.915 | <0.001*** |
| 1.0 cm | 5.117 | 0.778 to 9.457 | 0.124 | 2.503 | 0.021* | 35.796 | 7.517 to 64.076 | 0.098 | 1.742 | 0.014* |
| 1.5 cm | 4.613 | −0.241 to 9.467 | 0.101 | 1.990 | 0.062 | 33.021 | 1.768 to 64.275 | 0.073 | 1.253 | 0.039* |
| 2.0 cm | 1.775 | −3.038 to 6.588 | 0.062 | 1.183 | 0.464 | 12.801 | −17.208 to 42.810 | 0.019 | 0.313 | 0.397 |
| Mean | 1.799 | −1.951 to 5.550 | 0.069 | 1.724 | 0.343 | 13.104 | −10.243 to 36.450 | 0.032 | 0.695 | 0.268 |
| Maximum | 2.521 | 0.030 to 5.011 | 0.103 | 2.076 | 0.047* | 17.947 | 3.087 to 32.807 | 0.090 | 1.629 | 0.019* |
The measurements 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm refer to distances from the suture. Statistically significant models for EBL or EBL/kg with parietal bone thickness controlling for sex, age, and race via multivariate linear regression models are denoted as
p < 0.05,
p < 0.01,
p < 0.001.
When controlling for patient weight (kg), multivariate linear regression models revealed similar relationships between EBL per kilogram and parietal bone thickness 0.5 cm (p = 0.002) and 1.0 cm (p = 0.021) from the suture line, as well as the maximum parietal bone thickness (p = 0.047) (Table 5). It should be noted that these relationships had smaller effect sizes after accounting for patient weight at the time of surgery.
Parietal Bone Thickness and Transfusion
Patients who received intraoperative transfusion with PRBCs (n = 29, 23.4%) had significantly thicker parietal bone at 0.5 cm (2.331 mm vs 2.038 mm, p = 0.003), 1.0 cm (2.011 mm vs 1.776 mm, p = 0.17), and 1.5 cm (2.059 mm vs 1.856 mm, p = 0.021) from the suture line, but not at the suture (2.605 mm vs 2.299 mm, p = 0.068) or 2.0 cm from the suture line (2.043 mm vs 1.906 mm, p = 0.082) (Table 6 and Fig. 2). Patients who received intraoperative transfusion had no significant difference in the mean parietal bone thickness (1.917 mm vs 1.794 mm, p = 0.100) or maximum parietal bone thickness (3.897 mm vs 3.835 mm, p = 0.524). Patients who received a transfusion underwent surgery at ages comparable with those who did not receive a transfusion (3.58 months vs 3.53 months, respectively; p = 0.750).
Transfusion and parietal bone thickness
| Parietal Bone Thickness by Transfusion | Multivariate Logistic Regression for Transfusion | |||||||
|---|---|---|---|---|---|---|---|---|
| Transfusion | No Transfusion | p Value | β | 95% CI | SE | OR | p Value | |
| Suture line | 2.605 | 2.299 | 0.068 | 1.213 | −0.080 to 2.506 | 0.660 | 3.362 | 0.066 |
| 0.5 cm | 2.331 | 2.038 | 0.003** | 2.895 | 0.804 to 4.986 | 1.067 | 18.081 | 0.007** |
| 1.0 cm | 2.011 | 1.776 | 0.017* | 1.968 | 0.184 to 3.752 | 0.910 | 7.155 | 0.031* |
| 1.5 cm | 2.059 | 1.856 | 0.021* | 1.979 | 0.037 to 3.921 | 0.991 | 7.235 | 0.046* |
| 2.0 cm | 2.043 | 1.906 | 0.082 | 1.157 | −0.509 to 2.822 | 0.850 | 3.179 | 0.174 |
| Mean | 1.917 | 1.794 | 0.100 | 0.964 | −0.325 to 2.253 | 0.658 | 2.623 | 0.143 |
| Maximum | 3.897 | 3.835 | 0.524 | 0.284 | −0.532 to 1.100 | 0.416 | 1.328 | 0.496 |
The measurements 0.5 cm, 1.0 cm, 1.5 cm, and 2.0 cm refer to distances from the suture. Statistically significant models for intraoperative transfusion and parietal bone thickness controlling for sex, age, and race via multivariate logistic regression models are denoted as
p < 0.05,
p < 0.01.
Conditional estimate plots for transfusion and parietal bone thickness (mm) at the suture line (A) and 0.5 cm (B), 1.0 cm (C), 1.5 cm (D), and 2.0 cm (E) from the suture line, and for the mean bone thickness (F), maximum bone thickness (G), and age (months) at time of surgery (H). *p < 0.05; **p < 0.01.
Multivariate logistic regression models revealed findings with similar implications; parietal bone thickness 0.5 cm from the suture was most predictive of transfusion (Table 6). Patients with thicker parietal bone, 0.5 cm from the suture, were > 18 times more likely to require an intraoperative transfusion (OR 18.08, β 2.895, 95% CI 0.804–4.986; p = 0.007). Patients with parietal bones 1.0 cm and 1.5 cm from the suture line were > 7 times more likely to require intraoperative transfusion (OR 7.16, β 1.968, 95% CI 0.184–3.752; p = 0.031 and OR 7.24, β 1.979, 95% CI 0.037–3.921; p = 0.046, respectively). Multivariate logistic regression models did not significantly predict the need for transfusion based on the suture thickness (p = 0.066), parietal bone thickness 2.0 cm from the suture (p = 0.174), mean parietal bone thickness (p = 0.143), or maximum parietal bone thickness (p = 0.496). Univariate logistic regression models did not significantly predict the need for transfusion based on age at the time of surgery alone (p = 0.809).
Parietal Bone Thickness, LOS, and Complications
LOS was significantly associated with the mean parietal bone thickness (ρ = 0.243 and p = 0.018), but not with location-specific parietal bone thickness. In multivariate linear regression models controlling for age, sex, and race, the mean parietal bone thickness was predictive of LOS (β 0.575, 95% CI 0.095–1.054; p = 0.019).
Patients with complications (n = 10, 8.1%) had no significant differences in parietal bone thickness (all p > 0.05) (Table 3). Nine patients underwent reoperation, and this cohort had parietal bone thickness measurements similar to those who did not undergo reoperation (all p > 0.05).
Discussion
This study aimed to identify relationships between parietal bone thickness and perioperative outcomes, including EBL, blood transfusion rates, hospital LOS, and complications in patients who underwent SMC for nonsyndromic sagittal craniosynostosis. To our knowledge, relationships between bone thickness and perioperative or postoperative outcomes have not been evaluated, especially for patients undergoing invasive neurosurgery or craniofacial surgery procedures. Our findings suggest that increased cranial thickness may be related to perioperative outcomes in this patient population.
EBL was associated with parietal bone thickness, especially 0.5 cm to 1.5 cm from the suture line. These relationships were identified for EBL and EBL per kilogram on independent correlations and on multivariate linear regression models accounting for age, sex, and race. Our study also identified significant associations between transfusion rates and parietal bone thickness. Patients who received intraoperative transfusion had significantly thicker parietal bones, particularly 0.5 cm to 1.5 cm from the suture. Multivariate logistic regression models accounting for age, sex, and race predicted that patients with thicker parietal bones, 0.5 cm from the suture, were > 18 times more likely to require intraoperative transfusion, and patients with parietal bones 1.0 cm and 1.5 cm from the suture were > 7 times more likely to receive intraoperative transfusion. At our institution, craniectomy for SMC is typically 1.5 cm wide, or 0.75 cm from the midsuture line on either side.1 Notably, age at the time of surgery was not an independent predictor of these perioperative outcomes, and age was not independently associated with parietal bone thickness in this cohort of patients.
Literature on the surgical treatment of nonsyndromic craniosynostosis has demonstrated improved outcomes for patients undergoing SMC over other techniques, including cranial vault remodeling,6,7 craniectomy, and barrel staving,5 with one study demonstrating > 50% lower EBL with SMC compared with cranial vault remodeling.6 However, other studies have demonstrated longer intensive care unit durations and hospital LOSs compared with patients who underwent endoscopic strip craniectomy.3 The surgical complication rate in this study (8.1%) was also higher than the reported complication rates for patients who underwent endoscopic strip craniectomy, which has been reported as low as 2% to 3%.11 Our overall findings of transfusion rates and EBL are comparable with those of previous series,3,6 with a mean EBL of approximately 50 mL and intraoperative transfusion rate of 23.4%.
The results of this study, in the context of our institutional operative technique, could suggest that parietal bone thickness in craniectomy-specific locations is predictive of EBL and rates of transfusion. Thickened parietal bone could represent widening of the diploic space where physiological hematopoiesis is known to occur, rich with arteriovenous sinuses for blood flow.12 Thus, thicker parietal bone in the suturectomy area could lead to increased bleeding from areas with increased blood flow. Such findings may have implications for perioperative outcomes in patients undergoing other surgical procedures involving craniectomy. Because our cohort contained a group of same-procedure patients at a young age (younger than 1 year), it would be interesting to determine if the relationships described herein also persist in adult patients, or in patients undergoing craniectomy at different anatomical locations. Our findings could, however, be related to parietal bone thickness at sites of spring placement along craniectomy edges, in which case such findings may be specific to this particular patient cohort. The ability to generalize these findings may impact surgeon and family decision-making or provide additional information regarding anticipated transfusion needs and hospital LOS. It is worth noting that the hospital LOS is likely multifactorial, and additional work would be needed to determine clinical significance. Our data suggest that parietal bone thickness may serve as a more useful surrogate of perioperative outcomes compared with age alone, which raises questions regarding the role of preoperative parietal (or other calvarial bone) thickness analysis prior to invasive neurosurgery or craniofacial surgical intervention. Although morphological changes are beyond the scope of this paper, many newer surgical techniques, including SMC and distraction osteogenesis, rely on driving craniofacial morphological change through forces exerted to flex or bend bone. Understanding the bone characteristics responsible for these changes is essential, yet incomplete. Thus, gaining a more robust understanding of bone thickness, calcification, and other properties will be important in more completely describing morphological changes.
This study raises questions related to clinical management. Is there a practical role for preoperative CT thickness analysis for preferential osteotomy placement to thinner bone? Should the age range of patients who are offered minimally invasive stereotomy techniques be extended if parietal bone is thinner, despite older age? Similarly, could younger patients be considered surgical candidates, given comparable rates of EBL, transfusion, and complications? Reproducibility of these results, especially in other cohorts undergoing craniectomy or osteotomy for different indications and at different ages, will provide insights into the clinical relevance and applicability of the findings presented in the current study.
This study must be interpreted in the context of several limitations. The retrospective design precludes us from establishing causal conclusions, although reports of similar perioperative and clinical findings by previous studies on SMC in sagittal craniosynostosis substantiate our results presented here. Because our results hinge on parietal bone thickness analysis, limitations with image handling and quantification should be acknowledged. Additionally, there may be slight variation in image thresholding based on image quality and processing prior to import. Heat maps were applied to scans during thickness analysis to ensure that there were no areas of inadvertent misrepresentation of parietal bone thickness in order to mitigate any erroneous thickness measurements. We have also proposed significant relationships between blood loss and transfusion rates 0.5 cm to 1.5 cm from the suture in planned areas of suturectomy; however, the actual location of suturectomy in each patient was not confirmed with postoperative imaging. Additionally, this study was performed in a small cohort of pediatric patients who underwent a specific surgery for the same indication; thus, generalizability of these results to other cohorts or to older patients is unclear and may be limited. Moreover, we acknowledge the limitations of using EBL as a nonobjective perioperative outcome, yet believe transfusion rates with similar patterns of statistical significance substantiate these data. Finally, this study was primarily born out of exploratory analyses, with the primary aim of establishing relationships between parietal bone thickness, spring forces, and changes in the cranial index.
Despite these limitations, this study presents evidence for relationships between cranial thickness and perioperative outcomes in a cohort of patients who underwent craniectomy and SMC for nonsyndromic sagittal craniosynostosis. This study suggests that increased cranial thickness may be associated with increased intraoperative blood loss and need for transfusion, with the strongest evidence for increased cranial thickness in areas corresponding to suturectomy. Parietal bone thickness may serve as a useful surrogate of surgical safety and perioperative outcomes and could have important implications for patients undergoing other surgical procedures requiring craniectomy. Reproducibility of these results will be important to establish clinical significance. Further research on this topic may elucidate the relationships presented in this study and yield meaningful contributions to the care of our patients.
Conclusions
Parietal bone thickness may predict perioperative outcomes, including the need for transfusion, EBL, and hospital LOS in patients undergoing craniectomy and SMC for nonsyndromic sagittal craniosynostosis. Relationships between perioperative outcomes, especially the need for intraoperative transfusion and EBL, were most significant for parietal bone thickness 0.5 cm to 1.5 cm from the suture within the anticipated area of suturectomy, which was independent of age at the time of surgery.
Acknowledgments
This research was funded by the Division of Plastic, Reconstructive and Oral Surgery at the Children’s Hospital of Philadelphia and the Perelman School of Medicine at the University of Pennsylvania.
Disclosures
Dr. Swanson: consultant for KLS Martin and Synthes. Dr. Taylor: ownership in Ostiio.
Author Contributions
Conception and design: Tucker, Villavisanis, Cho, Shakir, Kalmar, Blum, Heuer, Madsen, Bartlett, Swanson, Taylor. Acquisition of data: Villavisanis, Shakir, Wagner, Cheung, Blum. Analysis and interpretation of data: Tucker, Villavisanis, Shakir, Taylor. Drafting the article: Villavisanis, Madsen. Critically revising the article: Tucker, Villavisanis, Cho, Kalmar, Lang, Bartlett, Swanson, Taylor. Reviewed submitted version of manuscript: Tucker, Villavisanis, Cho, Shakir, Kalmar, Wagner, Cheung, Blum, Lang, Heuer, Bartlett, Swanson, Taylor. Statistical analysis: Villavisanis.
References
- 1↑
Arko L IV, Swanson JW, Fierst TM, et al. Spring-mediated sagittal craniosynostosis treatment at the Children’s Hospital of Philadelphia: technical notes and literature review. Neurosurg Focus. 2015;38(5):E7.
- 2↑
David LR, Plikaitis CM, Couture D, Glazier SS, Argenta LC. Outcome analysis of our first 75 spring-assisted surgeries for scaphocephaly. J Craniofac Surg. 2010;21(1):3–9.
- 3↑
Lang SS, Meier PM, Paden WZ, et al. Spring-mediated cranioplasty versus endoscopic strip craniectomy for sagittal craniosynostosis. J Neurosurg Pediatr. 2021;28(4):416–424.
- 4
Lauritzen CGK, Davis C, Ivarsson A, Sanger C, Hewitt TD. The evolving role of springs in craniofacial surgery: the first 100 clinical cases. Plast Reconstr Surg. 2008;121(2):545–554.
- 5↑
Taylor JA, Maugans TA. Comparison of spring-mediated cranioplasty to minimally invasive strip craniectomy and barrel staving for early treatment of sagittal craniosynostosis. J Craniofac Surg. 2011;22(4):1225–1229.
- 6↑
Hassanein AG, Fadle KN. Assessment of the outcome of calvarial vault remodeling and spring-mediated cranioplasty in the correction of isolated sagittal suture synostosis. J Craniofac Surg. 2020;31(7):e747–e752.
- 7↑
Patel V, Shakir S, Yang R, et al. Perioperative outcomes in the treatment of isolated sagittal synostosis: cranial vault remodeling versus spring mediated cranioplasty. J Craniofac Surg. 2020;31(7):2106–2111.
- 8↑
Sun J, Ter Maaten NS, Mazzaferro DM, et al. Spring-mediated cranioplasty in sagittal synostosis: does age at placement affect expansion? J Craniofac Surg. 2018;29(3):632–635.
- 9↑
Wittner C, Borowski M, Pirl L, et al. Thickness accuracy of virtually designed patient-specific implants for large neurocranial defects. J Anat. 2021;239(4):755–770.
- 10↑
Love J, Selker R, Marsman M, et al. JASP: Graphical statistical software for common statistical designs. J Stat Softw. 2019;88(2):1–17.
- 11↑
Han RH, Nguyen DC, Bruck BS, et al. Characterization of complications associated with open and endoscopic craniosynostosis surgery at a single institution. J Neurosurg Pediatr. 2016;17(3):361–370.
- 12↑
Georgiades CS, Neyman EG, Francis IR, Sneider MB, Fishman EK. Typical and atypical presentations of extramedullary hemopoiesis. AJR Am J Roentgenol. 2002;179(5):1239–1243.
