Endoscopic-assisted management of sagittal synostosis: – wide vertex suturectomy and barrel stave osteotomies versus narrow vertex suturectomy

Abstract

Object

Endoscopic-assisted technique for treatment of craniosynostosis has reported benefits compared to open calvarial vault reconstruction. We evaluated two methods for endoscopic-assisted correction of sagittal synostosis: wide vertex suturectomy and barrel stave osteotomies (WVS+BSO) versus narrow vertex suturectomy (NVS).

Methods

We evaluated patients with nonsyndromic sagittal synostosis treated with either wide vertex suturectomy (4–6 cm) and barrel stave osteotomies (WVS+BSO) or narrow vertex suturectomy (NVS) (≈ 2 cm) from October 2006 to July 2013. Data prospectively collected included patient age, gender, operating time, estimated blood loss (EBL), postoperative hemoglobin (hgb), number of transfusions, complications, and cephalic index (CI). Fourteen patients in the NVS group were age matched to 14 patients in the WVS+BSO group. Descriptive statistics were calculated, and Student t-tests were used to compare prospectively obtained data from the WVS+BSO group with the NVS group in a series of univariate analyses

Results

Mean age at operative procedure was 3.9 months for WVS+BSO and 3.8 months for NVS. The mean operating time for patients undergoing NVS was 59.0 minutes, significantly less than the 83.4-minute operating time for patients undergoing WVS+BSO (p < 0.05). The mean EBL for patients undergoing NVS was 25.4 mL, which was not statistically different than the mean EBL for patients undergoing WVS+BSO – 27.5 mL. The mean postoperative hgb was 8.6 g/dL for the NVS group, which was not statistically different than the postoperative hgb of 8.0 g/dL for the WVS+BSO group. The mean preoperative CI was 69.9 in the NVS group, which was not statistically different than a CI of 68.2 in the WVS+BSO group. The average CI at one year of age was 78.1 in the NVS group, which was not significantly different than an average CI of 77.2 in the WVS+BSO group.

Conclusions

The NVS and WVS+BSO produced nearly identical clinical results, as cephalic index at one year of age was similar between the two approaches. However, the NVS required less procedural steps and significantly less operative time than the WVS+BSO. The NVS group obtained the final CI in a similar amount of time postoperatively as compared to the WVS+BSO group. Complications, transfusion rates, and EBL were not different between the two techniques.

Introduction

Numerous techniques have been developed to treat sagittal synostosis since Lane’s original description of the strip craniectomy in 1892.⁸ The endoscopic-assisted craniectomy (EAC) technique for treatment of craniosynostosis, first pioneered by Jimenez and Barone and reported in the literature in 1998³ has reported benefits compared to open calvarial vault reconstruction when performed in the first 6 months of life. The EAC technique has shown decreases in operative time, blood loss, transfusion requirements, cost, and hospital stays.^{1,3–6,11–13} Postoperatively, the infants wear custom-made cranial molding helmets to augment the reshaping of their calvaria.⁶

The EAC method for treatment of sagittal synostosis has two variations. The approach proposed by Jimenez and Barone relies on an endoscopic-assisted wide (4–6 cm) vertex craniectomy of the sagittal suture in addition to biparietal and bitemporal “barrel-stave” or wedge osteotomies.^3,5,7 A second approach reported by Ridgway et. al relies on less bone removal through an endoscopic-assisted narrow (≈ 2 cm) cm strip vertex craniectomy.^1,11 Both utilize postoperative cranial molding helmets and have shown effectiveness in treating sagittal synostosis. However, no study has directly compared both techniques. Here, we evaluated these two techniques for endoscopic-assisted correction of sagittal synostosis – wide vertex suturectomy and barrel stave osteotomies (WVS+BSO) versus narrow vertex suturectomy (NVS) – and compared variables associated with surgery and clinical outcomes.

Methods

Patients

This was an age-matched retrospective study with the data collected in a prospective manner. Fourteen patients with isolated sagittal synostosis underwent narrow vertex suturectomy (NVS) from January 2012 to July 2013. These patients were age-matched to 14 patients with isolated sagittal synostosis who had undergone wide vertex suturectomy and barrel stave osteotomies (WVS+BSO) at our institution from October 2006 to September 2011. Preoperative and postoperative clinical data have been prospectively collected at our institution since 2003. All patients underwent postoperative molding helmet therapy and were treated by a multidisciplinary team comprised of both a neurosurgeon (MDS) and plastic surgeon (either KBP or ASW). Data collected included patient age, gender, operating time, estimated blood loss (EBL), postoperative hemoglobin (hgb), number of transfusions, complications, and cephalic index (CI). Three-dimensional photographs assessed head shape preoperatively and postoperatively. The cephalic index was calculated as the ratio of the biparietal diameter to the AP diameter as measured by direct caliper measurements. The overall institutional goal cranial index for surgical repair of sagittal synostosis is 80. The Washington University institutional review board approved this study.

Surgical technique

Wide vertex suturectomy and barrel stave osteotomies (WVS+BSO)

The surgery was performed as described by Jimenez et al.^3,7 After general endotracheal tube anesthesia was established, the patient was placed in the modified “sphinx” position, frequently with the use of a Doro headrest (Pro Med Instruments, Freiburg im Breisgau, Germany). Two transverse incisions were made: a 2 cm incision posterior to the anterior fontanelle and a 2 cm incision anterior to lambda. After subgaleal dissection between the anterior fontanelle and lambda, bur holes and transverse osteotomies across the sagittal sinus were made at each incision using a high-speed drill and Kerrison rongeurs. The anterior osteotomy was connected to the anterior fontanelle with Mayo scissors, which allowed access to the epidural space with the endoscope. The endoscope was inserted into the epidural space to visualize dissection of the dura and sagittal sinus from the overlying fused suture. With the aid of the endoscope, a 4 – 6 cm wide vertex craniectomy was performed using Mayo scissors. Bilateral osteotomies are performed using scissors posterior to each coronal suture and anterior to each lambdoid suture, down to the squamosal suture inferiorly. The endoscope was used to visualize the cut bone edges to allow for coagulation using a suction/cautery device set at 50 Watts. After hemostasis, the skin was closed in layers using absorbable suture.

Narrow vertex suturectomy (NVS)

The surgery was performed as described by Ridgway et al.¹¹ The surgery was performed similarly to the WVS+BSO technique except a 2 cm wide vertex suturectomy and no bilateral barrel stave osteotomies were performed.

Helmeting

Helmeting remained the same with both techniques. Postoperatively, the child was fitted for his or her first in a series of 2 – 3 cranial molding helmets. The child wore the cranial molding helmets 23 hours per day. It was only taken off to clean the helmet and check the incision sites several times per day. Patients living locally used a helmet manufactured by a local orthotic provider. Patients who lived outside of the St. Louis metropolitan area were referred to local offices of a national orthotic provider (Hanger Orthotics) for molding helmet therapy. A nurse practitioner (S.N.) followed the patients for compliance every 1–3 months until termination of helmet therapy at approximately 1 year of age. Because cohorts were age matched and helmeting was continued until one year of age, there was no statistical difference in helmeting time. All patients received a 3D low-radiation-protocol head CT as an outpatient at one year of age.

Statistical Analysis

The 14 patients in the NVS group were age matched to 14 patients in the WVS+BSO group. Descriptive statistics were calculated, and Student t-tests were used to compare prospectively obtained data from the WVS+BSO group with the NVS group in a series of univariate analyses (Microsoft Excel 2010). The significance value was predetermined at p < 0.05.

Results

Patient Population

Fourteen patients underwent NVS from January 2012 to July 2013. Fourteen patients who underwent WVS+BSO from October 2006 to September 2011 were aged matched to the NVS group. The average age of the NVS group was 3.8 months (range: 2.1 – 6.5) and consisted of ten males and four females (Table 1). The average age of the WVS+BSO group was 3.9 months (range: 2.1 – 6.6) and similarly consisted of ten males and four females (Table 1).

Table 1.

Demographic characteristics and surgical and clinical outcomes for narrow vertex suturectomy (NVS) and wide vertex suturectomy and barrel stave osteotomies (WVS+BSO)

Characteristic	Number (%)
	WVS+BSO	NVS	p Value
Number of procedures	14	14	-
Age in months
mean	3.9	3.8	0.96
range	2.1–6.6	2.1–6.5	-
Gender (F/M)	4/10	4/10	-
Operative time (minutes)
mean	83.4	59	0.001
range	63–132	37–90	-
Estimated blood loss (EBL) (mL)
mean	28	25	0.60
range	10–60	15–40	-
Preoperative hemoglobin (g/dL)
mean	10.1	9.8	0.59
range	9–12.1	9–11.6	-
Postoperative hemoglobin (g/dL)
mean	8.0	8.6	0.24
range	5.5–11.9	7.2–10.4	-
Transfusions	1 (7)	0	0.33
Preoperative cephalic index (CI)
mean	68.2	69.9	0.27
range	63–75	62–78	-
Postoperative cephalic index (CI)
mean	77.2	78.1	0.60
range	72–86	73–85	-

Surgical Procedures

The 14 patients in the NVS group were compared to their age matched cohorts in the WVS+BSO group in terms of operating time, EBL, postoperative hgb, number of transfusions, and complications (Table 1, Fig. 1). The mean operating time for patients undergoing NVS was 59.0 minutes, significantly less than the 83.4-minute operating time for patients undergoing WVS+BSO (p < 0.05). The mean EBL for patients undergoing NVS was 25.4 mL, which was not statistically different than the mean EBL for patients undergoing WVS+BSO – 27.5 mL. The mean postoperative hgb was 8.6 g/dL (range: 7.2 – 10.4 g/dL) for the NVS group, which was not statistically different than the 8.0 g/dL (range: 5.5 – 11.9 g/dL) for the WVS+BSO group. No patient in the NVS group required a transfusion and one patient (7 %) in the WVS+BSO group required a postoperative transfusion due to a decreasing hgb level below 8 g/dL with the patient becoming symptomatic. There were no surgically related complications in either group.

Figure 1. Comparison of operative time, postoperative hemoglobin, and preoperative and postoperative cephalic index between WVS+BSO and NVS.

(A) Operative time was significantly decreased in the NVS group compared to the WVS+BSO group. However, the (B) postoperative hemoglobin and the (C) preoperative cephalic index and (D) postoperative cephalic index were not different between the groups. Error bars are standard deviation.

Surgical Outcomes

Each infant was evaluated and fit for a cranial molding helmet in the perioperative period and followed until one year of age. Cephalic index (CI) was obtained preoperatively and then postoperatively at one-month intervals until one year of age (Table 1, Fig. 1, Fig 2.). The mean preoperative CI was 69.9 in the NVS group, which was not statistically different than a CI of 68.2 in the WVS+BSO group (Table 1). The average CI at one year of age was 78.1 in the NVS group, which was not significantly different than an average CI of 77.2 in the WVS+BSO group. Just as the postoperative CI at one year of age were not different between the two groups, there was not a statistical difference in the postoperative monthly CIs between the two groups in the immediate months following surgery. The WVS+BSO group obtained the final CI in a similar amount of time postoperatively as compared to the NVS group (Fig. 2). Helmet therapy continued until one year of age for all patients in each cohort, and thus helmeting time was not statistically different between the groups. 3D CT representative examples of isolated sagittal synostosis before and after treatment with WVS+BSO and NVS are shown in Figure 3.

Figure 2. Percent of final cephalic index each month postoperatively until one year of age.

Percent of the final cephalic index (CI) as calculated by the difference between the cephalic index each month postoperatively and the final CI at one year of age divided by the final CI at one year of age. The WVS+BSO group obtained the final CI in a similar amount of time postoperatively as compared to the NVS group. One patient in the NVS was excluded as an outlier.

Figure 3. 3D CT representative examples of isolated sagittal synostosis before and after WVS+BSO and NVS.

(Left) Preoperative 3D CT and postoperative 3D CT at one year of age after WVS+BSO.

(Right) Preoperative 3D CT and postoperative 3D CT at one year of age after NVS.

Discussion

Sagittal synostosis has been treated using various surgical methods.^2,5,9,10 Comparing operative approaches is important to determine the best treatment strategy for each patient. Here we compared NVS and WVS+BSO to determine if there was any difference in operative and clinical outcomes. We found that mean operative times for NVS was 29 % less than the average operative times for WVS+BSO (p < 0.05). There was no difference in EBL, postoperative hgb, and complications between the two endoscopic assisted procedures. Both procedures achieved a similar clinical outcome upon completion of molding helmet therapy. The average CI at postoperative follow-up at one year of age for each method was nearly the same. No difference in CI each month postoperatively was noted between the WVS+BSO and NVS groups.

The difference in length of procedure between the two techniques was not unexpected since the NVS requires less tissue dissection and less bone removal as compared to the WVS+BSO. With only a 2 cm wide suturectomy, less subgaleal soft tissue dissection is needed in the NVS compared to the 4–6 cm wide suturectomy in the WVS+BSO. The WVS+BSO also utilizes multiple barrel-stave osteotomies. This requires more subgaleal soft tissue dissection in addition to the added time for barrel-stave osteotomies.

Interestingly, the EBL and postoperative hgb levels were not significantly different between the two methods. Although the WVS+BSO required more dissection and osteotomies, most bleeding occurred from the dura overlying the sagittal sinus and from the diploe of the bone after suturectomy. Therefore, the size of the suturectomy did not significantly affect blood loss. The majority of the blood loss for the procedure presumably ended once the dural bleeding was controlled and the bone edges cauterized.

The final clinical outcomes were unchanged between the two techniques. The CI for the NVS was similar to the CI for the WVS+BSO at one year of age. The WVS+BSO reached the final CI in a similar time frame as the NVS cohort. The bone removal utilized in the WVS+BSO is a substantial amount more than the NVS and therefore less redirected brain growth is needed postoperatively, but this did not seem to affect CI outcomes. In cases where the parents are compliant with follow-up appointments and conscientious about ensuring the child wear the helmet 23 hours a day, the outcomes between the two procedures appear to be the same. In both groups, the families and orthotists were counseled regarding aggressive helmeting therapy, and this was closely supervised by our center during the follow-up period to assure a good correction.

Limitations

This was a single center study and the numbers were small. Therefore, the results found here may differ from results found in comparable studies performed at other institutions. Although the data was collected prospectively, the analysis was conducted in a retrospective manner.

Conclusions

The NVS and WVS+BSO produced nearly identical clinical results, as cephalic index at one year of age was similar between the two approaches. However, the NVS required less procedural steps and significantly less operative time than the WVS+BSO. The WVS+BSO group obtained the final CI in a similar amount of time postoperatively as compared to the NVS group. Complications, transfusion rates, and EBL were not different between the two techniques.