Abstract

BackgroundTemporalis muscle (TM) atrophy is a commonly encountered cosmesis issue following craniotomies. This retrospective study aims to investigate the correlation of clinical and surgical factors with postoperative TM volume and head symmetry in patients undergoing frontotemporal craniotomy.

MethodsMedical records were retrospectively reviewed for patients undergoing pterional or middle fossa craniotomy. Pre- and post-operative clinical factors, intraoperative factors, and magnetic resonance imaging magnetic resonance imaging scans were obtained. TM volumes (cm³) were measured using volumetric analysis. 3-D reconstructed images of the head were independently reviewed to grade head symmetry on a binary scale. Pairwise correlation matrix, multiple linear regression (MLR), and logistic regression (LR) were used to assess the relationship of clinical and operative factors with TM volume and head symmetry.

ResultsA total of 46 patients were included for analysis, including pterional (n = 27) and middle fossa (n = 19) groups. The average TM volume loss was 36.8% (standard deviation [Std]: 17.18%) and 29.5% (Std: 16.72%) for patients receiving pterional and middle fossa craniotomies, respectively, indicating the pterional approach may carry a greater risk of disrupting the TM neurovascular supply ensuing subsequent cosmetic disfigurement. MLR identified age and scalp thickness as predictors of TM volume (P P

ConclusionPterional and middle fossa craniotomy procedures are both associated with significant TM volume loss and subsequent head asymmetry. Age and preoperative scalp thickness are independent risk factors for the development of TM volume loss and head asymmetry. These findings could contribute to the development of preoperative risk stratification algorithms, enabling neurosurgeons to predict the likelihood of cosmetic disfigurement. This would enhance preoperative counseling for patients undergoing craniotomy procedures, especially for elderly individuals and those with thin scalps. Overall, this study provides valuable insights into prognostic factors that may impact the development of poor cosmesis following frontotemporal craniotomy.

Keywords: Cosmesis, Head asymmetry, Pterional, Temporal hollowing, Temporalis muscle atrophy, Volumetric analysis

INTRODUCTION

Frontotemporal craniotomies are workhorse approaches frequently employed for access to the anterior and middle cranial fossae for diverse pathologies.[ 18 ] Their usage requires manipulation, elevation, and mobilization of the temporalis muscle (TM) and fascia, frequently resulting in postoperative TM atrophy and consequent cosmetic disfigurement.[ 6 , 14 ] Cosmetic outcomes, including TM atrophy and resultant asymmetry, significantly impact patients’ quality of life. Visible disfigurement, particularly on the face or head, can lead to psychosocial distress, including reduced self-esteem, social withdrawal, and symptoms of anxiety and depression. TM atrophy is defined as the wasting or reduction in size, volume, and mass of the TM, often due to disuse, denervation, injury, or underlying pathological conditions. The incidence of TM atrophy is estimated to be 87–100% following pterional craniotomies,[ 11 , 15 , 17 , 20 , 22 ] and the cause is typically secondary to direct dissection and reflection of the muscle,[ 3 ] electrocautery-induced disruption of its vascular supply,[ 9 ] or aggressive retraction leading to a loss of TM volume and subsequent asymmetry of the head.[ 3 , 6 , 8 , 10 ] Despite the recognition of TM atrophy as a potential cause of patient dissatisfaction following craniotomies, there exists a paucity of studies investigating the patient and surgery-specific factors that may increase the risk of TM atrophy.

The present study seeks to assess the relationship among preoperative and intraoperative factors, and subsequent postoperative TM volume and head symmetry in patients undergoing pterional craniotomy. Magnetic resonance imaging (MRI)-based volumetric analysis in three planes was used to estimate TM volume similar to previous studies.[ 10 ] MRI scans were then reconstructed to produce 3-D reconstructed images of the patient’s head, which was subsequently graded for symmetry. We hypothesized that clinical and intraoperative factors may serve as predictors of TM atrophy and changes in head symmetry postoperatively. By identifying correlates and predictors of TM atrophy and head asymmetry, this study will provide valuable insights for neurosurgeons in assessing the risk of cosmetic disfigurement following frontotemporal craniotomies.

MATERIALS AND METHODS

The International Review Board Committee approved this single-site retrospective study. Electronic medical records were obtained by searching the University of Missouri Hospital’s database for patients who underwent a pterional or middle fossa craniotomy. Patients who had prior craniotomies, bilateral craniotomies, < 18 years old, <2 months postoperative MRI since craniotomy or postoperative radiation therapy were excluded from the study.

Preoperative clinical factors, including patient demographics, body mass index (BMI), and smoking status, were collected. Using high-resolution T1 post-contrast MRI scans, we measured scalp thickness at the vertex and the midpoint of the ipsilateral and contralateral TMs in the coronal plane. Scalp thickness is an important factor to consider, given its role in mechanical protection from surgical manipulation, shared vascular supply through the superficial temporal artery (STA), and protection against scar tissue formation. In both pre-and postoperative MRI scans, we measured the volume (in cm³) from the top of the zygomatic process to the superior temporal line of the ipsilateral and contralateral TMs using iPlan (Brainlab, Munich, Germany).[ 4 ] Graphpad prism was used to do a one-way anova statistical test. This 3-D volumetric analysis was preferred over other methods due to the high-resolution images obtained from MRI, which enable high accuracy in locating the relevant anatomical regions of interest. We also obtained intraoperative variables, including procedure type (middle fossa or pterional), flap type (myocutaneous or separate scalp and muscle flaps), elevation technique (bovie or blunt), and total operation time. T1 post-contrast MRI scans were used to reconstruct 3-D images of each patient’s head.

Pairwise correlation matrix analysis was first performed to obtain Pearson correlation coefficients of clinical and intraoperative factors with postoperative TM volume and volume loss (two-tailed Student’s t-test). Multiple linear regression (MLR) was then performed to identify a linear model that can describe the impact of clinical and intraoperative correlates on postoperative TM volume following pterional craniotomies. Finally, an LR was performed to construct a model that can describe the relationship of clinical and intraoperative variables with the presence of head asymmetry. All statistical analyses and illustrations were performed in GraphPad Prism (version 10).

RESULTS

After the application of inclusion and exclusion criteria, the final cohort consisted of 46 patients [ Figure 1 ]. Twenty-seven and 19 patients who received pterional or middle fossa craniotomies, respectively, were included in this retrospective study. The clinical data are summarized in Table 1 . The average age of the cohort was 53 (range = 20–81), with an average BMI of 32 (range = 18–63). Most of the patients were non-smokers (n = 33, 72% of total). The average preoperative ipsilateral TM volume for all patients was 36cm³ (range = 15–67), with no difference between the pterional and middle fossa groups. Analysis of intraoperative factors revealed the most common flap type and elevation technique to be myocutaneous (n = 32, 70%) and bovie electrocautery (n = 36, 78%), respectively. The mean age was higher in the pterional group compared to the middle fossa group (57 vs. 46, P = 0.0422).

Figure 1:

Patient summary. Flow diagram for the selection of the study population.

Table 1:

Baseline characteristics of the included patients

TM volumetric analysis

Volumetric analysis of pre-and postoperative MRI head scans was performed as previously described [ Figure 2a ]. There was a significant reduction in ipsilateral TM volume (P < 0.001; 95% confidence interval [CI] 6.4–17.8) but not in contralateral TM volume [ Figure 2b ]. In addition, patients who received a pterional craniotomy had significantly lower TM volumes than patients who received a middle fossa craniotomy (P = 0.0034) [ Figure 2c ]. Following pterional and middle fossa craniotomies, the average percent loss of TM volume was 36.8% (Std = 17.2%) and 29.5% (Std = 16.7%), respectively.

Figure 2:

Volumetric analysis. (a) Temporalis muscles (TMs) were obtained (b) to assess TM atrophy and (c) to compare TM volume by craniotomy type.

MLR

A pairwise correlation analysis was first performed to identify correlates of postoperative TM volume and the change in volume. Age was negatively correlated (Pearson r = −0.414; P = 0.032), and ipsilateral scalp thickness positively correlated (Pearson r = 0.405; P = 0.036) with postoperative ipsilateral TM volumes [ Figure 3 ]. BMI positively correlated with the change in TM volume (Pearson r = 0.383; P = 0.049).

Figure 3:

Pairwise correlation matrix showing Pearson coefficient (in parentheses) and p-Value for correlations between variables (white = no correlation, light red = mild correlation, and dark red = moderate-high correlation).OR: Operating room.

These results were further modeled with MLR where age (P < 0.05) and ipsilateral scalp thickness (P < 0.05) correlated with ipsilateral TM volume in patients who received a pterional craniotomy [ Table 2 ].

Table 2:

Results of the multiple linear regression analysis showing predictors of ipsilateral TM volume following pterional craniotomy

Logistic regression (LR)

The best-fit LR model unveiled a positive correlation of age (r-squared = 0.4198) and ipsilateral scalp thickness (r-squared = 0.3449; P = 0.002) with head asymmetry probabilities, with a cutoff age of approximately 50 years [ Figure 4a ]. ROC analysis results unveiled that age, scalp thickness, and the change in ipsilateral TM volume predicted head asymmetry occurrence in the pterional group (area under the curve [AUC] = 0.91; AUC 95% CI 0.7921–1.000; P = 0.0053) [ Figure 4b ]. The positive predictive power was 86.36%, and the negative predictive power was 66.67%.

Figure 4:

Logistic regression. (a) Logistic regression modeling of increasing age impact on the predicted probability of head asymmetry (i.e., head symmetry = 0.0, head asymmetry = 1.0), and (b) ROC modeling of age, scalp thickness, and change in TM volume’s predictability of head asymmetry (binary). Red lines in (a): Age at which the individual is at 75% probability of head assymetry following pterional craniotomy. Pink line in (b): “Line of no-discrimination”. Points above the line indicate better-than-random performance. Points below the line suggest worse-than-random performance. Red dotted lines are defined as the age at which the individual is at a 75% probability of head assymetry. ROC: Response operating characteristics, TM: Temporalis muscle

DISCUSSION

Temporalis atrophy following craniotomy is a common postoperative complication where some degree of cosmetic disfigurement and functional compromise is probable. It is believed that atrophy of the TM is due to intraoperative factors such as (i) disruption of its neurovascular supply, (ii) inappropriate muscle tension, and/or (iii) direct injury to the muscle fiber.[ 16 ] In addition, atrophy can occur due to direct retraction of the deep temporal neurovasculature which course along the medial surface of the TM just superficial to the subperiosteum.[ 2 , 18 ] Many surgical techniques have been proposed to minimize TM atrophy, including (i) Intraoperative TM reconstruction,[ 21 ] (ii) Retrograde dissection,[ 16 ] (iii) Microfixation of the muscle using microscrews,[ 24 ] and (iv) Subfascial temporalis dissection through inferioposterior reflection.[ 9 ] Although these intraoperative strategies may provide benefits to the preservation of the TM function, cosmetic disfigurements such as head asymmetry and permanent scarring are still prevalent.

Current neuroimaging modalities used to estimate TM volume include computed tomography (CT), ultrasound (US), and, more recently, MRI.[ 1 , 10 , 12 ] Anand RK et al. used US to assess anterior TM volume as a general indicator for muscle hollowing in critically ill patients admitted into the intensive-care unit.[ 1 ] The US technique consisted of measuring the thickness or cross-sectional diameter along the length of the TM.[ 1 ] Kofler et al. used CT radiographs to estimate TM volume as a predictor of poor outcome variates in patients with subarachnoid hemorrhage.[ 12 ] Interestingly, the authors manually outlined the TM between the zygomatic arch and the inferior temporal line (in the axial plane), which produced a 3-D reconstructed volume of the muscle.[ 12 ] Hwang used MRI volumetric analysis similar to the present study to quantify the effects of electrocautery on TM volume in the axial plane only.[ 10 ]

In this retrospective study, we sought to quantify the degree of TM volume loss and identify preoperative and intraoperative predictive factors of postoperative TM volume using MRI volumetric analysis in three planes. We found that patients who received a pterional or middle fossa craniotomy had postoperative ipsilateral TM atrophy with an average pre- and postoperative ipsilateral volume of 36 (standard deviation [SD] = 11.4) and 24 (SD = 8.9) cm³, respectively, which represents an average of 36.8% reduction in volume. In pterional craniotomies, the volume of reflected TM is generally greater than in the middle fossa approach and may pose a greater risk of injury to the deep temporal neurovasculature and frontal branch of the facial nerve.[ 18 , 19 ]

Evaluation of preoperative and intraoperative factors with postoperative ipsilateral TM volume in the pterional group revealed that age and preoperative ipsilateral scalp thickness were negative and positive correlates, respectively. Previous reports by Lin et al.[ 15 ] found that age was negatively correlated with TM thickness (measured by CT angiography), which was an independent predictor of favorable functional independence (measured using Rankin score) in patients who received an endovascular thrombectomy. Our study is the first to demonstrate the negative outcome effects of increased age on postoperative TM volume using MRI 3-D reconstruction. A retrospective study conducted by Chung et al.[ 5 ] found thin scalps to be a predictor of poor wound healing in patients who received cerebral revascularization through the STA. In this procedure, blood is diverted from the scalp to the brain, causing necrosis and atrophy of the scalp, delaying proper wound healing mechanisms.[ 5 ] In the context of this study, thin preoperative scalps may indicate an STA more prone to hypoxia during TM reflection. The STA forms anastomoses with the deep temporal arteries; thus, if both are compromised, collateral circulation to preserve the TM may be impaired, resulting in significant scarring and/or TM atrophy, causing head symmetry defects.

Cosmetic dissatisfaction is a common complaint following pterional craniotomies. Specifically, hollowing in the mid-temporal area of the head, located medial to the inferior margin of the frontal bone, is associated with the use of electrocautery and procedure dissatisfaction.[ 15 ] Although we were not able to test the impact of electrocautery on head symmetry, we did observe significant hollowing in the mid-temporal area most pronounced in those receiving pterional craniotomies. Due to the absence of cosmetic complaints in some cases,[ 23 ] there are likely other factors that contribute to increased atrophy of the muscle. For example, Grajeda-García and Mercado-Caloca[ 7 ] found that using an osteomyoplastic flap instead of a myocutaneous reduced the degree of TM hollowing in patients receiving pterional craniotomies. Other factors, such as procedure duration[ 11 ] and surgical technique, play a role.[ 13 ] In this study, we found that increasing age and ipsilateral scalp thickness increases the risk of head symmetry defects. In this study, we found that patients at 40 years of age have a 50% likelihood of head asymmetry, 75% at ~50 years of age, and 90–100% at >65 years of age. This threshold highlights the importance of considering patient age as a significant risk factor during preoperative counseling and planning, as it may guide discussions on cosmetic expectations and the potential need for adjunctive interventions to mitigate asymmetry in older patients.

This retrospective study highlights age and ipsilateral scalp thickness as factors linked to postoperative TM atrophy and predictors of head asymmetry. This study will require further validation using a larger cohort and a more even distribution of intraoperative factors to identify actionable predictors of TM atrophy and poor cosmesis. Some of these factors may include surgical techniques, such as the extent of soft-tissue dissection, size of skin incision, and methods used for muscle preservation or reattachment; procedure duration, as prolonged operative times could exacerbate tissue stress and ischemia; the use and extent of electrocautery, which may contribute to muscle fibrosis and atrophy through thermal damage; reconstruction methods, such as comparing osteomyoplastic versus myocutaneous flaps, which could influence the degree of tissue support and cosmetic outcomes; and patient-specific variables, including baseline muscle thickness, scalp elasticity, comorbidities such as diabetes or vascular disease that impair healing, and previous surgical history. This will also enable researchers to assess the interactions between preoperative factors, including scalp thickness and intraoperative factors. Further study will enable researchers to construct deep learning artificial-intelligence models such as multi-layer perceptrons and deep neural networks to simultaneously utilize linear and non-linear associations between variables to make more accurate predictions of TM atrophy and head asymmetry. These studies will require larger datasets with a more even distribution of variables across craniotomy groups to effectively link pre- and intra-operative factors to TM atrophy and cosmetic outcomes. These findings will provide valuable insights for neurosurgeons, otolaryngologists, and plastic surgeons to enable surgeons to customize their approach to mitigate unfavorable outcomes for patients.

Limitations

The limitations of this study stem primarily from its retrospective design, which inherently entails certain limitations in data collection and analysis. Retrospective studies rely on existing data, which may be incomplete or subject to inherent biases in documentation. In our study, despite efforts to minimize bias through rigorous data collection and analysis, the retrospective nature limits our ability to control for all potential confounding variables. In addition, the sample size of our study, although adequate for statistical analysis, may limit the generalizability of our findings to broader populations. Furthermore, while we employed MLR and LR analyses to identify predictors of TM atrophy and head asymmetry, there may exist other unmeasured variables or interactions between variables that could influence these outcomes.

CONCLUSION

Frontotemporal craniotomy procedures are linked to notable TM volume reduction and consequent head asymmetry. Age and scalp thickness before surgery are predictors for TM volume loss and asymmetry of the head. This research offers valuable perspectives on prognostic factors that may influence the occurrence of unsatisfactory cosmesis following frontotemporal craniotomies. These findings have important clinical implications, particularly in guiding preoperative patient counseling. Identifying patients at higher risk for TM atrophy and head asymmetry allows surgeons to provide tailored discussions regarding potential cosmetic outcomes, ensuring informed consent and realistic expectations. In addition, integrating these prognostic factors into surgical planning could help refine approaches, such as minimizing aggressive muscle retraction or employing alternative techniques to preserve TM integrity. Furthermore, the study underscores the need for developing standardized protocols aimed at reducing TM atrophy. These protocols might include optimizing surgical techniques, enhancing postoperative rehabilitation, or investigating novel interventions to mitigate muscle volume loss. Implementing such measures could improve both cosmetic and functional outcomes, contributing to greater patient satisfaction and quality of life.

Ethical approval

The research/study approved by the Institutional Review Board at University of Missouri-Columbia School of Medicine, number 2093653, dated October 11, 2023.

Declaration of patient consent

Patient’s consent was not required as there are no patients in this study.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Use of artificial intelligence (AI)-assisted technology for manuscript preparation

The authors confirm that there was no use of artificial intelligence (AI)-assisted technology for assisting in the writing or editing of the manuscript and no images were manipulated using AI.

Disclaimer

The views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of the Journal or its management. The information contained in this article should not be considered to be medical advice; patients should consult their own physicians for advice as to their specific medical needs.

Acknowledgments

The authors would like to thank their institution for granting access to view, collect, and process clinical data included in this article.

References

1. Anand RK, Bhattacharjee A, Baidya DK, Subramaniam R, Rewari V, Ray BR. Feasibility of anterior temporalis muscle ultrasound for assessing muscle wasting in ICU: A prospective cohort study. J Ultrasound. 2023. 26: 653-61

2. Bae H, Choi YJ, Lee KL, Gil YC, Hu KS, Kim HJ. The deep temporal arteries: Anatomical study with application to augmentations procedures of the temple. Clin Anat. 2023. 36: 386-92

3. Bowles AP. Reconstruction of the temporalis muscle for pterional and cranio-orbital craniotomies. Surg Neurol. 1999. 52: 524-9

4. Brainlab AG. iPlan cranial [computer software]. Available from: https://www.brainlab.com/surgery-products/overview-neurosurgery-products/cranial-planning [Last accessed on 2025 Apr 03].

5. Chung J, Lee S, Park JC, Ahn JS, Park W. Scalp thickness as a predictor of wound complications after cerebral revascularization using the superficial temporal artery: A risk factor analysis. Acta Neurochir (Wien). 2020. 162: 2557-63

6. Goncalves DB, Dos Santos MI, De Cristo Rojas Cabral L, Oliveira LM, Da Silva Coutinho GC, Dutra BG. Esthetics outcomes in patients submitted to pterional craniotomy and its variants: A scoping review. Surg Neurol Int. 2021. 12: 461

7. Grajeda-García FM, Mercado-Caloca F. El colgajo osteomioplástico, una contribución al arte de la neurocirugía [The osteomyoplastic flap, a contribution to neurosurgery]. Rev Med Inst Mex Seguro Soc. 2011. 49: 649-54

8. Honeybul S. Management of the temporal muscle during cranioplasty: Technical note. J Neurosurg Pediatr. 2016. 17: 701-4

9. Horimoto C, Toba T, Yamaga S, Tsujimura M. Subfascial temporalis dissection preserving the facial nerve in pterional craniotomy--technical note. Neurol Med Chir (Tokyo). 1992. 32: 36-7

10. Hwang SW, Abozed MM, Antoniou AJ, Malek AM, Heilman CB. Postoperative temporalis muscle atrophy and the use of electrocautery: A volumetric MRI comparison. Skull Base. 2010. 20: 321-6

11. Kim E, Delashaw JB. Osteoplastic pterional craniotomy revisited. Neurosurgery. 2011. 68: 125-9 discussion 129

12. Kofler M, Reitmeir P, Glodny B, Rass V, Lindner A, Ianosi BA. The loss of temporal muscle volume is associated with poor outcome in patients with subarachnoid hemorrhage: An observational cohort study. Neurocrit Care. 2023. 39: 198-206

13. La Rocca G, Della Pepa GM, Sturiale CL, Sabatino G, Auricchio AM, Puca A. Lateral supraorbital versus pterional approach: Analysis of surgical, functional, and patient-oriented outcomes. World Neurosurg. 2018. 119: e192-9

14. Lee SW, Lee YS, Lee MS, Suh SJ, Lee JH, Kim JW. Cosmetic outcome after electrocautery versus non-electrocautery dissection of the temporalis muscle for pterional craniotomy. J Cerebrovasc Endovasc Neurosurg. 2022. 24: 16-23

15. Lin YH, Chung CT, Chen CH, Cheng CJ, Chu HJ, Chen KW. Association of temporalis muscle thickness with functional outcomes in patients undergoing endovascular thrombectomy. Eur J Radiol. 2023. 163: 110808

16. Oikawa S, Mizuno M, Muraoka S, Kobayashi S. Retrograde dissection of the temporalis muscle preventing muscle atrophy for pterional craniotomy. Technical note. J Neurosurg. 1996. 84: 297-9

17. Park J, Hamm IS. Cortical osteotomy technique for mobilizing the temporal muscle in pterional craniotomies. Technical note. J Neurosurg. 2005. 102: 174-8

18. Rodriguez Rubio R, Chae R, Vigo V, Abla AA, McDermott M. Immersive surgical anatomy of the pterional approach. Cureus. 2019. 11: e5216

19. Salgado-Lopez L, Perry A, Graffeo CS, Carlstrom LP, Leonel L, Driscoll CL. Anatomical step-by-step dissection of complex skull base approaches for trainees: Surgical anatomy of the middle fossa approaches and anterior petrosectomy, surgical principles, and illustrative cases. J Neurol Surg B Skull Base. 2022. 83: e232-43

20. Santiago GF, Terner J, Wolff A, Teixeira J, Brem H, Huang J. Post-neurosurgical temporal deformities: Various techniques for correction and associated complications. J Craniofac Surg. 2018. 29: 1723-9

21. Spetzler RF, Lee KS. Reconstruction of the temporalis muscle for the pterional craniotomy. Technical note. J Neurosurg. 1990. 73: 636-7

22. Thiensri T, Limpoka A, Burusapat C. Analysis of factors associated with temporal hollowing after pterional craniotomy. Indian J Plast Surg. 2020. 53: 71-82

23. Youssef AS, Ahmadian A, Ramos E, Vale F, Van Loveren HR. Combined subgaleal/myocutaneous technique for temporalis muscle dissection. J Neurol Surg B Skull Base. 2012. 73: 387-93

24. Zager EL, DelVecchio DA, Bartlett SP. Temporal muscle microfixation in pterional craniotomies. Technical note. J Neurosurg. 1993. 79: 946-7