Reappraising the bifrontal decompressive craniectomy: A systematic literature review of indications, outcomes, and clinical applications

Abstract

Background:

Bifrontal decompressive craniectomy (BDC) is used to manage elevated intracranial pressure (ICP) from bilateral frontal swelling caused by severe traumatic brain injury (TBI), malignant cerebral edema, or other acute intracranial pathologies, but conflicting outcome data have been reported. This study aimed to evaluate the efficacy of BDC across various etiologies, identify gaps to guide research, and outline a framework for clinical decision-making.

Methods:

A systematic search of MEDLINE, PubMed, Scopus, Embase, and Cochrane databases was conducted. In addition, 3 formalin-fixed, noninjected cadaveric head specimens were dissected to illustrate BDC and its anatomical correlates.

Results:

Of 410 articles, 54 met inclusion criteria (742 patients). TBI was the leading etiology (656 of 742 patients; 88.4%), notably from motor vehicle collisions (58.4%) and falls (19.7%). On initial examination, 315 of 519 patients (60.7%) had a Glasgow Coma Scale score ≤8. Mean (standard deviation) time from injury to BDC was 2.5 (1.9) days, intensive care unit stay was 12.9 (5.0) days, hospital stay was 66.7 (38.9) days, and time to cranioplasty was 83.4 (40.7) days. Glasgow Outcome Scale values were reported for 494 patients: 205 patients, 5; 75 patients, 4; 71 patients, 3; 24 patients, 2; and 119 patients, 1.

Conclusion:

BDC may offer a survival benefit in patients with otherwise-fatal elevations in ICP, but cases require more detailed tracking. Functional outcomes vary, and the risks of developing severe disability are substantial. Implementing BDC in acute neurocritical care requires ethically grounded patient selection and family counseling, incorporating prognosis, patient values, and shared decision-making.

INTRODUCTION

Bifrontal decompressive craniectomy (BDC) is a potentially life-saving neurosurgical intervention employed in the management of bilateral frontal contusions, malignant cerebral edema, and other acute intracranial pathologies. BDC involves removing a bifrontal bone flap to alleviate elevated intracranial pressure (ICP).[13,16] Primarily intended to prevent imminent brain herniation, BDC allows symmetric brain expansion, recovery of cerebral circulation, and, when required, a wide surgical corridor for addressing bilateral frontal pathologies, such as hematomas, necrosis, infection, or tumors with extensive perilesional edema.[8,39,45,49,58-60,76]

The procedure was initially documented by Miyazaki et al.[50] in Japan in 1966, subsequently gaining recognition within the English scientific community through the work of Kjellberg and Prieto[45] in 1971, as a definitive intervention for the management of refractory intracranial hypertension. Early descriptions referred to this procedure as a “craniotomy,” emphasizing the intention of eventual bone flap replacement. Contemporary practice classifies it as a “craniectomy,” or the removal of bone without immediate replacement, providing space for cerebral expansion and diminishing the risk of secondary ICP-related injury.[8,60,68]

The utility, technique, and implementation of BDC have evolved over the past 5 decades, predominantly reserved for cases of otherwise-fatal intracranial hypertension unresponsive to conservative management.[4,68] The urgent clinical situations in which BDC is performed often hinder meticulous, data-driven documentation, thereby contributing to a paucity of robust evidence to guide its application and resulting in variable outcomes. Current guidelines for BDC are primarily based on clinical judgment or extrapolated from studies involving other forms of decompressive craniectomy. Consequently, the specific indications, optimal timing, extent of bony resection, and patient outcomes remain inadequately defined.

This study reviews the current evidence regarding the clinical indications, patient eligibility, surgical techniques, patient outcomes, and the risk-benefit profile of BDC across various pathologies to facilitate clinical decision-making. Ethical considerations are addressed, and technical descriptions are illustrated with cadaveric dissections to underscore operative nuances.

MATERIALS AND METHODS

Search strategy

Systematic searches of the MEDLINE, PubMed, Scopus, Embase, and Cochrane Library databases were conducted on February 11, 2026, without time restriction, using the following terminology: [(bifrontal decompressive craniectomy), (bifrontal craniectomy), (bicoronal craniectomy), (bicoronal decompressive craniectomy), (bifrontal craniectomies), (bifrontal decompressive craniectomies), (bicoronal craniectomies), (bicoronal decompressive craniectomies), (bifrontal DC), (bicoronal DC)]. Boolean operators “AND” and “OR” were employed to combine these medical subject heading terms, thereby ensuring comprehensive coverage. Formatting modifications were implemented to align the search query with the advanced search criteria of each database.

Eligibility criteria

Eligible studies for screening encompassed case reports, case series, retrospective and prospective cohorts, and randomized controlled trials (RCTs). Only publications in English were considered. In cohorts containing multiple surgical techniques or conservative control groups, only the subset treated with BDC was included. In addition, only cases that (1) explicitly stated the preoperative clinical indication, (2) detailed the BDC procedure adequately, and (3) provided postoperative outcomes were incorporated. Nonoriginal studies such as reviews, editorials, commentaries, animal and cadaveric studies, errata, retractions, and conference proceedings were excluded.

Study selection

The retrieved studies were systematically filtered based on their titles and abstracts, with duplicates subsequently excluded. An additional search, including manual review and reference checking, was performed to identify further relevant studies. Two independent reviewers (J.T.O. and K.Y.) conducted the screening process utilizing the Rayyan Platform, in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.[26,55] Any disagreements were resolved through discussion, and consensus was reached, resulting in a total of 54 articles included in the final analysis [Figure 1].

Figure 1:

Flow diagram showing the study selection process. BDC: Bifrontal decompressive craniectomy. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Quality assessment

Risk-of-bias assessments were conducted on all included full-text articles using the Joanna Briggs Institute critical appraisal tools.[3] Checklists pertinent to case reports, case series, cohorts, and RCTs were employed. These tools were selected for their systematic methodology in evaluating internal validity and relevance, with criteria specifically adapted to each study design. Quality assessments were independently performed by 2 reviewers (J.T.O. and K.Q.), with any conflicts resolved through review by 2 additional reviewers (K.Y. and J.E.P.).

Cadaveric study

Three formalin-fixed, noninjected cadaveric head specimens were dissected to illustrate the relevant surgical anatomy of the BDC procedure and to contextualize the technical nuances described in the literature over time. Burr hole placement, the extent of bony removal, and 2 commonly used durotomy techniques were illustrated [Figures 2-4].

Figure 2:

Cadaveric dissection to demonstrate the skin incision and reflection of the myocutaneous flap. The specimens were positioned supine and secured in a Mayfield 3-point rigid fixation system, with the head flexed 15°–30° and kept in a neutral (nonrotated) position. (a-c) The location of the bregma, and therefore the coronal suture (CS), was estimated at approximately 12.5 cm from the nasion along the midsagittal line. The skin incision was marked approximately 2 cm posterior to the CS. The short purple dotted line indicates the location of the CS. The long purple dotted line indicates the location of the skin incision, which extends bilaterally to the zygomatic root, just anterior to the tragus. (d-f) The incision was performed with particular attention to preserving the superficial temporal artery (STA). The cutaneous flap was reflected anteroinferiorly over the face. The dashed line denotes the CS. (g-i) Bilateral temporalis muscles (m.) were reflected inferolaterally. The dashed line denotes the CS. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Figure 4:

Cadaveric dissection to demonstrate durotomy. Following maximal bony removal, 2 specimens were used to illustrate alternative durotomy techniques. (a-c) In the first specimen, bilateral cruciate dural incisions were marked, deliberately avoiding the superior sagittal sinus (SSS). [29] (d-f) Cruciate dural incisions were performed, and the dura was reflected. (g-i) In the second specimen, a bilateral semicircular dural incision was marked. (j) Ligatures were placed 1 cm apart on the anteriormost part of the exposed SSS. (k-m) A semicircular dural incision was performed between the ligatures, and the dura was reflected. (n) To theoretically demonstrate maximal decompression, the SSS and falx were then divided with scissors. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Three-dimensional rendering

Three-dimensional (3D) reconstructions were produced using the 3D Slicer (https://www.slicer.org/) and Blender (https://www.blender.org/) software applications to enhance visualization of the resection boundaries, the dimensions of the excised bone flap during craniectomy, and various surgical modifications [Figure 5].

Figure 5:

Three-dimensional rendering of the bony exposure to demonstrate options for burr hole placement and craniectomy. (a) Axial view showing the location of the parietal burr holes if drilling adjacent to the superior sagittal sinus (SSS) using the 2-parietal-burr-hole technique. Parietal burr holes should be placed 1 cm posterior to the coronal suture (CS; black dashed line). The red dashed lines indicate the borders of the craniectomy. The yellow dashed lines indicate the borders of the optional “bucket handle” craniectomy. (b) Sagittal view showing the location of the bilateral MacCarty and squamous temporal burr holes, using the 2-parietal-burr-hole technique. (c) Axial and (d) sagittal views showing the “bucket handle” craniectomy for preservation of the SSS using the 2-parietal-burr-hole method. (e) Axial view showing the location of the single parietal burr hole if drilling at midline, through which the SSS is visualized. (f) Sagittal view showing the location of the bilateral MacCarty and squamous temporal burr holes, using the single-parietal-burr-hole technique. (g) Axial and (h) sagittal views showing the full bifrontal decompressive craniectomy bony exposure (for maximal decompression), which can be obtained using either the single- or 2-parietal-burr-hole methods. Bone segmentations were performed in 3D Slicer (https://www.slicer.org/) by tracing key cranial sutures for anatomical accuracy. Final 3D rendering and visualizations were completed in Blender (https://www.blender.org/). Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

RESULTS

Study selection

Of the 410 articles retrieved, 54 met the inclusion criteria: 39 case reports, 8 case series, 6 retrospective cohorts, and 1 RCT [Figure 1, Table 1[1,5,6,10-12,15,17,18,20-23,25,28,31,36,40,41,43,46-48,51,52,54,56,61-67,69,75,77,78,83] and Table 2[4,13,16,19,33,39,45,49,58,59,70,72,76,79,82]].

Table 1:

Summary of individually reported cases of patients who underwent BDC^a.

Table 2:

Summary of case series, retrospective and prospective cohorts, or randomized controlled trials of patients who underwent BDC.^a

Quality assessment

The methodological quality was evaluated for all selected studies. Quantitative risk of bias was determined based on the percentage of methodological quality questions answered affirmatively by the reviewers: 70% or higher denoted a low risk, 50%–69% represented a moderate risk, and 49% or less signified a high risk of bias. Among the 54 studies included, 45 (83%) were classified as low risk, 8 (15%) as moderate risk, and 1 (2%) as high risk. The predominant concerns pertained to inconsistent or incomplete data reporting or undisclosed study methodologies [Supplemental Table 1].[13,22,41,51,58,61,65,70,72]

Historical and geographical distribution

Since its introduction in 1971, the frequency of BDC implementation has fluctuated. The next reported use of BDC was in 1975, followed by a 20-year gap until it became more widespread in 1995. BDC became increasingly popular thereafter, and the number of annual publications gradually increased, reaching up to 5 per year between 2007 and 2025 [Figure 6a]. The use of BDC and its associated patient outcomes has been reported by 22 countries since 1971. The countries reporting implementation included the United States (11 of 54 studies; 20%), Australia (9 of 54 studies;17%), and Germany (5 studies; 9%), among others, including 1 multicenter RCT held in Australia, Saudi Arabia, and New Zealand [Figure 6b].

Figure 6:

Annual number and geographic distribution of publications reporting bifrontal decompressive craniectomy. (a) Number of publications per year from 1995 to 2025. There were also publications in 1971 and 1975 (1 each), for a total of 54 publications. (b) Geographical distribution of the included 54 articles. The United States (11 studies; 20%), Australia (9 studies; 17%), and Germany (5 studies; 9%) were involved in the most studies. China, India, Italy, and Saudi Arabia were each involved in 3 (6%) of the studies. Korea, Canada, Switzerland, and New Zealand were each involved in 2 (4%) of the studies. The United Kingdom, Peru, Spain, Singapore, Pakistan, Mexico, Iran, Greece, Brazil, Taiwan, and Jordan were each involved in 1 (2%) of the studies. The multicenter randomized controlled trial involving Australia, Saudi Arabia, and New Zealand was counted once for each participating country. Map created with Datawrapper (https://www.datawrapper.de/). Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Patient presentation and initial work-up

A total of 742 patients who underwent BDC were included in the study, consisting of 510 (68.7%) men, 219 (29.5%) women, and 13 (1.8%) patients of unspecified sex. The mean age (standard deviation [SD]) of the 599 patients with reported age was 34.1 (11.6) years.

The primary indication for BDC was traumatic brain injury (TBI) (656 of 742 patients; 88.4%). Traumatic mechanisms of injury included motor vehicle collisions (383 of 656 patients; 58.4%), falls (129 of 656 patients; 19.7%), all-cause penetrating head wounds explicitly specified as the primary mechanism (32 of 656 patients; 4.9%), and assault or combat injuries with an unspecified primary mechanism (20 of 656 patients; 3.0%), among others. Nontraumatic etiologies (86 of 742 patients; 11.6%) encompassed tumors (34 of 86 patients; 39.5%), immune-mediated or inflammatory processes (9 of 86 patients; 10.5%), ruptured aneurysms (5 of 86 patients; 5.8%), and other uncommon pathologies manifesting as diffuse cerebral edema [Figure 7a].

Figure 7:

Patient characteristics. (a) Patient etiologies that indicated bifrontal decompressive craniectomy (BDC) included 656 traumatic brain injury (TBI) and 86 non-TBI mechanisms. The 656 TBIs (left) included 383 (58.4%) motor vehicle collisions (MVCs), 129 (19.7%) falls, 83 (12.7%) unspecified TBIs (N/S), 32 (4.9%) all-cause penetrating head wounds (PHWs) specified as the primary mechanism, 20 (3.0%) assault or combat injuries with an unspecified mechanism, (A/C) and 9 (1.4%) crush injuries (Crush). The 86 non-TBIs (right) included 34 (40%) tumors (32 frontal meningiomas, 1 sinus nasal undifferentiated carcinoma, and 1 pituitary macroadenoma), 9 (10%) immune-mediated or inflammatory processes (IMIPs), including multiple infectious agents and encephalopathies; 5 (6%) ruptured aneurysms involving the anterior communicating, basilar, and ophthalmic arteries; and 38 (44%) other pathologies, such as diabetic ketoacidosis, disseminated frontal osteomyelitis, and hyperammonemic encephalopathy. (b) Patients with presenting Glasgow Coma Scale (GCS) scores were categorized according to the standard head trauma classification, including a further subdivision of the severe category. Of 519 patients, 90 (17.3%) were classified as severe with GCS scores of 3–5, 136 (26.2%) as severe with GCS scores of 6–8, 89 (17.1%) as severe with GCS scores of 3–8 but not further specified (not shown in the figure), 138 (26.6%) as moderate with GCS scores of 9–12, and 66 (12.7%) as mild with GCS scores of 13–15. (c) Mean (SD) durations are shown for 317 patients with reported time from injury to BDC, 197 patients with reported intensive care unit (ICU) stay duration, 215 patients with reported total hospital stay duration, and 115 patients with reported postoperative interval until cranioplasty (CP). (d) Patient outcomes reported using Glasgow Outcome Scale (GOS) scores at the most recent follow-up in each report. Of 531 patients, 205 (38.6%) had a good recovery (GOS score 5), 75 (14.1%) had moderate disability (GOS score 4), 71 (13.4%) had severe disability (GOS score 3), 24 (4.5%) were left in a vegetative state (GOS score 2), and 119 (22.4%) died (GOS score 1). The remaining 37 (7.0%) patients with GOS scores given in a range but not further specified (8 with GOS scores 1–3 and 29 with GOS scores 4–5) are not shown in the figure. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

The 519 patients who had a Glasgow Coma Scale (GCS) score at initial examination were stratified according to the head-trauma classification as mild (GCS score 13–15), moderate (GCS score 9–12), and severe (GCS score 3–8), enabling us to descriptively assess outcomes in relation to the initial severity of neurological impairment. Among these 519 patients, 66 (12.7%) presented with mild head trauma, 138 (26.6%) presented with moderate head trauma, and 315 (60.7%) presented with severe head trauma. The cohort of patients with severe head trauma was further divided into those with GCS scores of 6–8 (136 of 519; 26.2%), those with GCS scores of 3–5 (90 of 519; 17.3%), and those with GCS scores reported as 8 or below but with no further stratification provided (89 of 519; 17.1%) [Figure 7b].

The preoperative radiographic characteristics were obtained through computed tomography or magnetic resonance imaging and documented for 620 patients, each demonstrating several relevant findings indicative of intracranial hypertension [Table 3].

Table 3:

Summary of radiographic characteristics reported for patients undergoing bifrontal decompressive craniectomy.

Patient management

Conservative medical therapies for managing elevated ICP were reportedly attempted in 466 of 469 (99.4%) patients for whom pre-BDC therapeutic methods were documented. Furthermore, other case-specific surgical interventions performed before BDC included the insertion of an external ventricular drain in 84 of 469 patients (17.9%), alternative trephination procedures (such as burr hole or craniotomy evacuation of lesions, or other decompressive craniectomy) in 11 patients (2.3%), aneurysm coiling or clipping in 5 patients (1.1%), and placement of a ventriculoperitoneal shunt in 1 patient (0.2%). It is important to note that interventions performed concurrently with or subsequent to BDC were not included in these data.

Data were available regarding the interval between injury and BDC for 317 patients, intensive care unit (ICU) stay for 197 patients, total hospital length of stay for 215 patients, and the interval between BDC and cranioplasty for 115 patients. The mean (SD) time to BDC was 2.5 (1.9) days after injury (or following the determination of pathology). The mean (SD) durations of ICU stay and total hospital stay were 12.9 (5.0) days and 66.7 (38.9) days, respectively. The mean (SD) time to cranioplasty (after BDC) was 83.4 (40.7) days [Figure 7c].

Outcomes

The Glasgow Outcome Scale (GOS), including the GOS-Extended (GOS-E), was the most frequently reported outcome measure. In studies employing GOS-E, values were converted to the standard GOS to ensure consistency [Supplemental Table 2].[80] Independent case reports [Table 1] and case series or cohort studies [Table 2] lacking an objective outcome measure for the entire cohort or using non-GOS parameters were excluded from further quantitative analysis.

A postoperative or follow-up GOS score was documented for 531 of 742 patients (71.6%). At the most recent assessment within each report, 205 patients (38.6%) achieved a good recovery (GOS score 5), 75 patients (14.1%) exhibited moderate disability (GOS score 4), 71 patients (13.4%) experienced severe disability (GOS score 3), 24 patients (4.5%) remained in a vegetative state (GOS score 2), and 119 patients (22.4%) had died (GOS score 1). For the remaining 37 patients (7.0%), the GOS score was reported as a range, with 29 attaining a favorable outcome (GOS score 4–5) and 8 an unfavorable outcome (GOS score 1–3); further stratification was not provided in the respective studies [Figure 7d]. Among the independent case reports and cohort studies lacking an objective outcome metric, at the most recent evaluation, 180 of the remaining 211 patients (85.3%) were alive, 10 (4.7%) were in a vegetative state, and 21 (10.0%) had died.

Eleven articles documented both the initial GCS score and the subsequent GOS score, although the outcomes were reported at different postoperative time points. A comparison of these outcomes across the studies indicated an improvement in patients’ recovery over time [Table 4].[4,5,16,20,21,33,39,47,58,76,82]

Table 4:

Available GCS and GOS data among varying outcome evaluation time points.

DISCUSSION

The role of BDC in clinical practice is constrained primarily by a lack of evidence and conflicting outcome data. The 2020 Brain Trauma Foundation guidelines, although not specifically focused on the bifrontal approach, provide recommendations for decompressive craniectomy in refractory cases but leave significant gaps.[27] The expanding body of evidence and the increasing adoption of the procedure necessitate a comprehensive understanding of the critical values that may inform clinical decision-making.

Indications and management timeline

A broad spectrum of etiologies may result in refractory intracranial hypertension, thereby serving as indications for BDC [Figure 7a]. The procedure is predominantly employed for bilateral frontal pathologies or injuries that induce swift neurological deterioration. Although BDC has demonstrated efficacy in the rapid reduction of ICP, uncertainties persist concerning the optimal timing and surgical technique specific to each clinical scenario.[76,79]

Following the staircase model, initial management of acute ICP elevation commences with conservative measures, such as head elevation, fluid resuscitation, ventilatory support, sedation, analgesia, neuromuscular blockade, or administration of hyperosmotic agents (e.g., mannitol and urea).[57,68] Additional case-specific options before BDC include the placement of an external ventricular drain and the evacuation of focal lesions through a burr hole or craniotomy.

As demonstrated in this review, first-tier therapies are often insufficient to control ICP, necessitating “tier-three therapies” (TTTs), such as barbiturates, therapeutic hypothermia, or decompressive craniectomy. A recent study investigating the real-world applications of TTTs concluded that decompressive craniectomy, although not specifically BDC, is often performed earlier and more frequently than other TTTs and is associated with improved neurological outcomes and decreased mortality.[57] Commentary on this article agrees, emphasizing that early BDC may be appropriate and beneficial in select cases.[53]

Certain conservative treatments, such as barbiturate therapy and therapeutic hypothermia, pose inherent risks when administered continuously or over extended periods, potentially exacerbating outcomes due to adverse effects and cumulative toxicity.[53,57,71] Moreover, delays in pursuing decompressive craniectomy may diminish its efficacy.[53] Multiple studies highlight the significance of operative timing as a crucial factor influencing patient outcomes.[20,79] Polin et al.[58] observed favorable results in 13 of 28 patients when BDC was performed within 48 h, whereas no positive outcomes were recorded beyond that timeframe. Zhaofeng et al.[82] employed an operation-time scale to monitor patient status and advocated for timely intervention, asserting that BDC should be conducted before the occurrence of secondary brain damage. This increasing focus on the importance of operative timing is encouraging; however, surgical technique and other variables are also likely to impact long-term results.

Surgical technique

Since their inception in 1971, BDC techniques have exhibited variability across studies, ostensibly due to the acute clinical environment and the diverse extent of injury or pathology. The standard procedure commences with the patient in a supine position, with the head maintained in a neutral or slightly flexed alignment.[29,45,82] Most of the researchers describe a bicoronal skin incision that extends bilaterally just anterior to the zygomatic roots and posterior to the coronal suture, thereby facilitating the anterior reflection of a substantial cutaneous flap. Subsequently, the bilateral temporal muscles are reflected inferolaterally [Figure 2].[45]

Although efforts have been made to delineate the optimal extent of bony resection, existing evidence endorses a craniectomy window adequate for alleviating intracranial hypertension and, if required, facilitating access to intracranial pathologies. The scope of bone removal exhibits considerable variation throughout the literature, encompassing diverse burr hole placement strategies as well as conservative and extensive decompressive procedures [Figure 3].[8,9,11,14,16,17,19,24,25,29,33,36,39,45,51,58-62,65,66,72,73,75,76,79,82]

Figure 3:

Cadaveric dissection to demonstrate burr hole placement and craniectomy. Once full bony exposure was achieved, the borders for the planned craniectomy were carefully delineated. All figures demonstrate the estimated location of the superior sagittal sinus (SSS; purple dotted line) as a critical landmark throughout the dissection. (a-c) The first specimen underwent a 2-parietal-burr-hole technique. Burr holes were placed 1 cm lateral to midline and 1 cm posterior to the coronal suture (CS), straddling the SSS. Burr holes were created bilaterally at the MacCarty keyhole and in the squamous temporal region near the coronal plane of the parietal burr holes. (d-f) Demonstration of the “bucket handle” as a bilateral craniectomy technique for protection of the SSS using the 2-parietal-burr-hole method. The preserved bony strip measures approximately 2 cm in width, extending 1 cm on either side of the SSS. (g-h) The second specimen used a single midline parietal burr hole technique. A burr hole was placed 1 cm posterior to the CS. Burr holes were created bilaterally at the MacCarty keyhole and in the squamous temporal region near the coronal plane of the parietal burr hole. (j-l) Following careful separation of the sinus and dura from the bone using a Penfield dissector, a footplate was used to complete the craniectomy. Both specimens achieved maximal exposure by connecting the MacCarty keyholes along the supraorbital ridge and the temporal and parietal burr holes 1 cm posterior to the CS in a continuous line to create a single, removable bone flap. Used with permission from Barrow Neurological Institute, Phoenix, Arizona.

Parietal burr holes have been positioned at a distance of 1–2 cm anterior and 3–6 cm posterior to the coronal suture.[4,12,16,19,33,58] In a coronal plane, multiple studies endorse the placement of bilateral parietal burr holes approximately 0.5–1 cm from the midline to preserve options that mitigate the risk of injuring the superior sagittal sinus (SSS).[16,29,60] However, the technique involving 2 parietal burr holes requires meticulous and often blind dissection of the SSS from the bone. Conversely, because expansion of the brain parenchyma may impede safe exposure, some authors advocate for a single midline parietal burr hole directly over the SSS, facilitating earlier and enhanced visualization for prompt repair in the event of sinus injury due to trauma or iatrogenic factors.[9,14,45] Concerning the inferolateral borders, nearly all studies employed the MacCarty keyhole and an additional burr hole in the squamous temporal bone aligned in the same coronal plane as the parietal burr hole(s). Nonetheless, 1 study uniquely utilized the temporal line as the lateral boundary, considering it sufficient for decompression.[82]

The border of the frontal craniectomy is conventionally delineated along or immediately superior to the supraorbital ridge. In certain instances, this approach may result in exposure of the frontal sinus, a topic addressed comprehensively in numerous studies. The prevailing recommendation is to avoid entry into the sinus whenever feasible; notwithstanding, cranialization is recommended to diminish the risks associated with cerebrospinal fluid leaks, pneumocephalus, or infections.[11,16,17,66,75,82] Conversely, some experts contend that avoiding sinus exposure, particularly in cases involving mass lesions, may constrict the operative corridor and necessitate more aggressive brain retraction, thereby elevating the likelihood of intraoperative complications.[59]

Ideally, the bone flap should be excised in a single piece to facilitate rapid and maximal decompression while preserving the possibility for subsequent cranioplasty. Alternatively, a 2-segment bone flap maintains a strip of bone over the SSS in a “bucket handle” configuration, thereby optimizing its protection.[4,29,36,61,79] This method has been considered in cases of bilateral frontal contusions with moderate ICP elevation, where limited decompression appears adequate. Furthermore, if additional decompression becomes necessary, various studies describe the removal of portions of the bilateral parietal bones, squamous temporal bones, and greater wings of the sphenoid.[16,33,51,61]

Following bony resection, durotomy is recommended to enhance decompression. However, the potential risks of brain prolapse and unnecessary cortical injury have led to debate regarding its routine application.[51] Various durotomy techniques have been described in the literature. In several studies, a large semicircular durotomy was created crossing the midline, requiring suture ligation and division of the anteriormost SSS and falx.[8,39,45,58,60,62,65,72] Others emphasized the risk of hemorrhage and opted to preserve the SSS, employing bilateral cruciate, stellate, or fish-mouth durotomies [Figure 4].[16,76] Although not demonstrated in our dissections, synthetic dura, periosteal, or temporalis fascia grafts may be applied to cover the durotomy in select cases.[25,61,72,79] To date, no consensus has been established concerning the optimal placement of burr holes, the extent of craniectomy, or the choice of durotomy technique; rather, these decisions are individualized based on the location, type, and severity of the underlying pathology.

Cranioplasty is performed following clinical stabilization to restore cranial integrity, improve cosmetic appearance, and reestablish cerebrospinal fluid dynamics and cerebral perfusion.[36] Decades ago, scholars recommended early cranioplasty or the reimplantation of an unsecured bone flap after decompressive craniectomy to facilitate prompt restoration of cranial integrity. Although these approaches offer an immediate postoperative protective buffer, subsequent studies have discouraged this practice due to concerns about impaired decompression and elevated complication rates.[13,16,73,79] Current clinical practice increasingly favors delayed reconstruction to optimize the procedure’s objectives, although the influence of the timing of cranioplasty on patient outcomes remains a topic of debate.

Outcomes

Numerous cohorts and case studies indicate that BDC provides a mortality advantage in cases of severe TBI and ruptured aneurysms, demonstrating effectiveness in promptly reducing ICP.[17,33,49,65,76,79,82] Nonetheless, severe TBI frequently involves multiple injuries that extend beyond the therapeutic capabilities of BDC. Consequently, the effectiveness of BDC should be assessed within the broader context of the overall injury or disease severity.

Despite its interrater variability and limited diagnostic specificity, the GCS score remains a valuable baseline metric for patient selection, prognostication, and outcome assessment through the GOS score. Analysis across various studies has identified key trends but also underscored limitations to relying exclusively on the GCS for prognostication. Consistent with earlier research, a low GCS score is a strong predictor of poor outcomes and mortality, with some studies completely excluding such patients.[39,47] However, several large cohorts that included patients with severe GCS scores reported reasonably favorable GOS outcomes, indicating that early neurological compromise does not necessarily preclude recovery, and BDC may benefit even severely injured patients [Table 4].[58,76]

Intermediate (9–12) or high (13–15) baseline GCS scores may indicate increased recovery potential, and many authors endorse considering these patients as more appropriate candidates for BDC.[4,82] Studies with large samples of intermediate GCS scores exhibited the greatest variability in recovery. Although favorable outcomes were common, the initial GCS score alone did not reliably predict successful recovery [Table 4].[4,20,24,33] This variability likely reflects the influence of operative timing, surgical techniques, adjuvant therapies, rehabilitation, cranioplasty, and complications. Notably, 1 study demonstrated the significance of long-term assessment, because patient GOS scores improved between 6 and 18 months of follow-up, emphasizing that recovery from TBI and major surgery can be gradual and nonlinear.[33] Consequently, early assessments, as observed in many other studies, may underestimate potential long-term outcomes.

The superiority of BDC over conservative management in diffuse TBI remains a subject of debate. The Decompressive Craniectomy (DECRA) trial stands as the only clinical trial to date that has directly evaluated the efficacy of BDC in comparison with conservative management.[16] The findings indicated that early BDC, using nonexhaustive conservative therapies, was associated with comparable mortality rates but poorer functional outcomes relative to standard medical treatment alone.[16] These results contradicted earlier literature and the initial hypothesis of the trial. Proposed explanations for these findings included axonal stretch caused by brain expansion, alterations in cerebral blood flow and metabolism, and potential surgical complications. Despite its significance in a field characterized by a lack of prospective data, the DECRA trial faced methodological criticisms and did not definitively resolve the ongoing debate regarding patient selection and treatment strategies.[2,34,35,38,44] A more recent investigation, known as the Randomized Evaluation of Surgery with Craniectomy for Uncontrollable Elevation of Intracranial Pressure (RESCUEicp), examined the application of decompressive craniectomy in cases of late refractory ICP elevation.[37] In this review, RESCUEicp did not report data on the bifrontal approach in comparison with other forms of decompressive craniectomy. Nonetheless, the results demonstrated a promising 22% reduction in mortality at 6-month follow-up; however, there was an increase in the proportion of patients experiencing severe disability or in a vegetative state relative to medical management alone. Collectively, these trials laid the foundational evidence underpinning the current recommendations issued by the Brain Trauma Foundation.[27] Beyond their conclusions regarding mortality and functional outcomes, the results of both studies supported the role of decompressive craniectomy in controlling ICP and reducing ICU stay duration and emphasized the importance of patient selection and operative timing in influencing outcomes.[16,32]

Complications

Various complications associated with BDC have been documented in the academic literature. Nonetheless, there remains debate regarding whether these complications stem from the decompression procedure itself or from the primary brain injury or underlying pathology. The most frequently observed complications include subdural or subgaleal hygromas or effusions, cortical herniation, seizures, hydrocephalus, and infections.[32] Additional significant complications encompass cerebrospinal fluid leakage, intracranial hematoma, epidural fluid accumulation, diffuse axonal injury, and sepsis, among others.[17,20,25,32,33,36,41,51,58,75,82] Furthermore, an inadequately sized bone flap may result in cortical damage or hemorrhagic infarction due to brain herniation through the craniectomy defect. After early cranioplasty, there is also an associated risk of strangulation.[24,30,32,41,51,61,79,81]

Another significant but infrequent complication is the “syndrome of the trephined,” which can lead to neurological deficits, headaches, seizures, mood swings, and behavioral disturbances. The syndrome often poses diagnostic challenges and may be confused with other conditions due to its rarity. Fortunately, the syndrome of the trephined is frequently alleviated through cranioplasty.[8,41] As BDC garners more defined indications and acceptance, it is incumbent upon clinicians to identify the potential complications in relation to the patient’s comprehensive health status.

Ethical considerations

Since its inception, BDC has been valued for mitigating the secondary consequences of elevated ICP in otherwise-fatal scenarios.[58] Therefore, any reduction in mortality is generally viewed as a clinical improvement. However, recent literature highlights the ethical and prognostic complexities of decompressive craniectomy. Despite the associated reduction in mortality, many providers hesitate to perform BDC due to concerns about poor functional outcomes, unnecessary suffering, and increased medical burden.[39] The procedure’s nuanced risk-benefit profile complicates the development of evidence-based guidelines that support appropriate patient selection.[6,7,36,39] At a minimum, decisions should consider the patient’s comorbidities, age, baseline independence, severity of illness, and patient or family preferences.[6] Patients and providers alike must be educated on the potential for persistent disability, emotional lability, and cognitive deficits as well as the added risk of anxiety, depression, and suicidality seen in survivors compared with the general population.[15] These challenges underscore the importance of careful patient selection, comprehensive family counseling, postoperative rehabilitation, and coordinated interdisciplinary care, all of which are essential for long-term follow-up.[6,18,33,41]

Limitations of the included studies and future directions

The included articles encompass a period of 5 decades, representing a significant advancement in neurotrauma care, which limits the comparability of their findings. The observed efficacy of BDC may be overstated due to survivor bias, because reported outcomes predominantly reflect patients who survived to complete follow-up and may underrepresent patients with unfavorable outcomes not documented in the literature. Although it was a priority to preserve the integrity of the original literature, data were manually extracted and occasionally required minor adjustments in terminology or presentation to ensure clarity and consistency.

The individual case reports described unique clinical scenarios involving BDC but often did not focus on evaluating the procedure’s efficacy. As a result, these reports often lacked objective patient metrics, such as GCS score, ICP level, operative timing, or follow-up GOS score. Furthermore, several studies, including case series, cohort studies, and the DECRA trial, presented data in aggregated or descriptive formats, which limited insight into individual patient trajectories and precluded a quantitative meta-analysis.

Future studies should provide standardized, case-level information to improve data quality. Although thorough documentation may seem secondary in critical scenarios, optimizing the utility of BDC depends on longitudinal, meticulous, data-driven analysis. The ongoing challenge of identifying appropriate candidates for BDC and defining its clinical use encourages further research involving key variables, including the underlying etiology, preoperative neurological and radiological findings, ICP measurements, operative timing, interval to cranioplasty, and multiple long-term neurological assessments.

CONCLUSION

BDC can be a vital yet inadequately examined neurosurgical procedure for addressing malignant intracranial hypertension resulting from diverse etiologies. To the best of our knowledge, this study is the first and most comprehensive review dedicated exclusively to BDC, providing a succinct synthesis of the evolving literature concerning its technique, clinical applications, and ethical considerations. Despite its theoretical advantages, BDC is infrequently employed, as reflected in the limited literature, likely due to a complex risk-benefit profile and the relative rarity of diffuse frontal pathologies. Data from this review show that BDC is typically performed in patients with severe neurologic compromise (GCS score <8) and imaging signs of impending brain herniation. Although BDC may reduce mortality and facilitate functional recovery in selected patients, severe disability or death continue to be prevalent. The variability in outcomes likely reflects differences in indications, injury mechanisms, clinical presentations, and timing of treatment. These findings endorse more detailed patient data tracking, transparent prognostic counseling, ethical decision-making, standardized reporting of outcomes, and increased utilization of functional and quality-of-life assessments. Despite these limitations, BDC persists as a viable option for bilateral frontal pathologies associated with emergent or refractory intracranial hypertension.

Ethical approval:

Institutional Review Board approval is not required.

Declaration of patient consent:

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

Financial support and sponsorship:

This study was supported by the Newsome Chair in Neurosurgery Research held by Dr. Preul and funds from the Barrow Neurological Foundation.

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.