REVIEWARTICLE - BRAIN INJURY

Mini-craniotomy as the primary surgical interventionfor the treatment of chronic subdural hematoma—a retrospectiveanalysis

Jorn Van Der Veken & Johnny Duerinck & Ronald Buyl &Katrijn Van Rompaey & Patrick Herregodts &Jean D’Haens

Received: 21 November 2013 /Accepted: 16 February 2014 /Published online: 11 March 2014# Springer-Verlag Wien 2014

AbstractBackground The incidence of chronic subdural hematoma(CSDH) is increasing, but optimal treatment remains contro-versial. Recent meta-analyses suggest burr hole (BH) drainageis the best treatment because it provides optimal balancebetween recurrence and morbidity. Mini-craniotomy may of-fer supplementary technical advantages while maintainingequal or better outcomes. This study investigates theoutcome of mini-craniotomy as the sole treatment inpatients with CSDH.Method We analyzed all patients operated on for CSDH withmini-craniotomy in our neurosurgical center between 2005–2010. Baseline patient characteristics (age, sex, comorbidities,imaging characteristics, known risk factors for developmentof CSDH and neurological examination at presentation) andoutcomes (mortality, complications, recurrence and neurolog-ical examination at discharge) were recorded.Results One hundred twenty-six adult patients were included,mean age was 73.9 (range 18 to 95) years old, and the sex ratio(M:F) was 2:1. Eighty-four percent of the patients showedclinical improvement at discharge, as shown by a decrease intheMarkwalder score postoperatively (with 57%Markwalder

0 and 23 % Markwalder 1). Recurrence rate was 8.7 %.Overall complication rate was 34.1 % (27.8 % medical com-plications and 6.3 % surgical complications). In-hospital mor-tality was 13.5 % (8.7 % due to pulmonary infections and1.6 % to surgical complications). Preoperative Markwaldergrade correlated significantly with complication rate, as didthe presence of a neurodegenerative disease (p=0.018). Fac-tors significantly related to mortality in univariate analysiswere arterial hypertension (p=0.038), heart failure (p=0.02),renal failure (p=0.017), neurodegenerative disease (p=0.001),cerebrovascular accident (p=0.008) and coagulopathy (p=0.019). Multivariate analysis was not able to confirm anysignificant relationship.Conclusion This is the first published series of CSDH inwhich all consecutive patients were operated on by mini-craniotomy. The invasiveness and complication rate of mini-craniotomy are equal to those of burr hole treatment, butvisualization is superior, resulting in lower recurrences. Arandomized controlled trial is indicated to identify the bestsurgical strategy for the treatment of CSDH.

Keywords Chronic subdural hematoma .Mini-craniotomy

Introduction

Chronic subdural hematoma (CSDH) is one of the most com-mon pathologies encountered by neurosurgeons. The incidenceof CSDH increases proportionally with age, with a recentestimated incidence of 20.6/100,000/year for patients between70 and 79 years old and 127.1/100,000/year in the over 80 agegroup [11]. Older studies reported incidences of only 7.35-8.2 /100,000/year in the age group older than 65 years [2, 8]. Due tothe increasing average age of the Western population and the

Portions of this work were presented in poster form at the AnnualScientific Meeting of the Belgian Society of Neurosurgery in Ghent,Belgium

J. Van Der Veken (*) : J. Duerinck :K. Van Rompaey :P. Herregodts : J. D’HaensDepartment of Neurosurgery, Universitair Ziekenhuis Brussel,Brussels, Belgiume-mail: jpvdveke@vub.ac.be

R. BuylDepartment of Biostatistics and Medical Informatics, VrijeUniversiteit Brussel, Brussels, Belgium

Acta Neurochir (2014) 156:981–987DOI 10.1007/s00701-014-2042-8

fact that CSDH mainly affects the elderly, a further increase inincidence is expected. This emphasizes the necessity of guide-lines concerning optimal treatment.

Even though it is one of the most frequently encounteredcranial pathologies in neurosurgical practice, much controver-sy still remains regarding the best surgical treatment. Pub-lished results are highly variable, with recurrence rates rang-ing from 0.35 % to 60 % [22, 31]. Complication rates from0 % to 34 % have been reported while surgical mortality instudies ranges from 0 % to 24 % [17, 21, 24, 28].

Recent meta-analyses propose burr hole (BH) drainage asthe most efficient choice because it provides the best balancebetween recurrence and morbidity [14, 23, 29]. It should benoted, however, that the pooled data used for these meta-analyses are very heterogeneous and often biased toward themost commonly used treatment modality. The meta-analysesinvestigating the craniotomy technique are mostly based onsmall populations and are often biased since they are not con-trolled or randomized [7, 14, 30]. Moreover, most studies andmeta-analyses do not make the distinction between large crani-otomy and mini-craniotomy, where the diameter of the boneflap remains limited to 3-4 cm [12]. Mini-craniotomy is lessoften used nowadays, probably as a result of these publications.

To our knowledge, there are no published surgical serieswhere mini-craniotomy is used as the only technique for thetreatment of CSDH in adult patients. This technique offerstheoretical advantages over other treatment options, such as abetter visualization of the subdural space and the possibility towidely open membranes, while the claimed disadvantages oflonger operating time and higher complication rate couldprove to be less prominent when only considering mini-craniotomy and disregarding large craniotomy.

Methods

Patient selection and data

In our neurosurgical center, mini-craniotomy performed undergeneral anesthesia is the preferred technique for treatment ofCSDH. We retrospectively analyzed all consecutive patientsoperated on for CSDH in our center between January 2005and December 2010 and compared these results with those inthe literature. Data were obtained by reviewing neurosurgicallogbooks and patient files at the University Hospital UZBrussel, Brussels, Belgium.

Baseline patient characteristics such as age, sex, comorbid-ities, known risk factors for development of CSDH (use ofanticoagulants or antiplatelet drugs, presence of a coagulationdisorder, a ventriculo-peritoneal shunt and a history of alcoholabuse) and neurological examination at presentation wereregistered. Comorbidities that were listed were arterial hyper-tension (AHT), diabetes mellitus (DM), heart failure (HF),

renal failure (RF), chronic obstructive pulmonary disease(COPD), the presence of a neurodegenerative disease (ND),hyperlipidemia and a prior cerebrovascular accident (CVA).Neurodegenerative diseases consisted of dementia andParkinson’s disease. History of alcohol abuse was defined asthe consumption of four or more alcoholic beverages per day.

Pre- and postoperative CT and/or MRI images were ana-lyzed. The type of CSDH was classified as either mixeddensity, homogeneous and isodense to the brain or homoge-neous and hypodense to the brain. The neurologic perfor-mance of the patients was evaluated pre- and postoperativelyusing the ‘Markwalder Neurological Grading System,’ whichis the most commonly used neurological grading system forCSDH [16] (Fig. 1).

Clinical and radiological status of the patients after treat-ment were evaluated upon discharge from the hospital. Dateof latest follow-up was recorded. Uniform criteria were usedto define mortality, complications and recurrence rate. Mor-tality was defined as death reported between surgery anddischarge from the hospital. Complications included all med-ical and surgical complications other than recurrence or mor-tality during hospitalization. Surgical complication was acomplication directly related to the surgical intervention. Re-currence was defined as reoperation for an ipsilateral symp-tomatic CSDH within 6 months after the previous surgicalevacuation (reoperation rate). Hospital stay was defined as theduration from the initial operation to discharge. In reoperatedcases, the duration of hospitalization from the second opera-tion onwards was also added to the hospital stay.

Surgical technique

In the literature, craniotomy is defined as an opening of theskull with a diameter larger than 30 mm and with reposi-tioning of the bone flap at the end of the intervention. Inmini-craniotomy, the diameter of the bone flap remains limit-ed to 3-4 cm. Openings with a diameter between 5 and 30 mmare called burr holes (BH), those smaller than 5 mm twist drillcraniostomy [30].

All surgeries in this series were performed under generalanesthesia. The small osteoplastic craniotomy was performedin most cases using a trephine. In ten patients, the mini-craniotomy was performed by connecting one or two burrholes with the craniotome. After the craniotomy the dura

Fig. 1 Markwalder grading score

982 Acta Neurochir (2014) 156:981–987

had been incised, the outer membrane of the hematoma coag-ulated and was maximally opened. Following the initial re-moval of the hematoma or remaining clots, the subdural spacewas irrigated rigorously with physiological saline water untilthe drained fluid became clear. A subdural Jackson Prattdrainage catheter was then placed, the bone flap replaced, andthe skin closed. The Jackson-Pratt catheter was left in situ for aminimum of 24 h and a maximum of 96 h, its removal depend-ing on the amount of drained volume of blood and the result ofpostoperative CT or MRI imaging. In case of bilateral hemato-mas, the patient was included in the study as one case, and bothsides were treated simultaneously using the same technique.

Statistical analysis

Variables considered in the statistical analysis included sex,age, hypertension, diabetes mellitus, chronic obstructive pul-monary disease, heart failure, renal failure, neurodegenerativedisease, parkinsonism, cerebrovascular accidents, ethylism,antiplatelet therapy, anticoagulant therapy, coagulopathy andhematoma characteristics.

For statistical analysis, IBM SPSS (v. 20) software wasused. We performed a univariate analysis to assess the rela-tionship between each variable and the outcome (mortality,recurrence, complications) by applying the Mann-Whitney Utest (for non-categorical) and the chi-square test (for categor-ical variables). All significant univariate results were thentested in a multivariate logistic regression model.

All tests were performed using a 5 % significance level.

Results

Patient characteristics

During the 6-year evaluation period, 131 patients underwentsurgery for CSDH. Five patients were excluded from theanalysis because they were younger than 18 years. Eighty-six (68 %) patients were male, and 40 (32 %) were female.Average age was 73.4 years old, 73.2 in the male group and75.3 in the female group, with a range of 18 to 95 years old.Age and gender distributions of the patient population areillustrated in Fig. 2.

104 out of 126 (84.6 %) patients had at least one medicalcomorbidity (see "Materials and methods" for listing). In 75patients (59 %), there was a history of trauma, and 4 patientshad a ventriculo-peritoneal shunt. Almost half of the patients(56/126; 44 %) were treated with anticoagulants (n=20) orantiplatelet agents (n=36). Seven patients (5 %) had a coag-ulopathy. Eighteen patients (14 %) had a history of chronicalcohol abuse. Thirty-six out of the 75 (48 %) patients whohad a history of trauma were on anticoagulant medication or

antiplatelet agents versus 20 out of 47 (42 %) patients whereno clear cause could be found (Table 1).

Symptoms and signs at presentation are shown in Fig. 3.The most frequent clinical finding at presentation was motordeficit in 36 patients (28 %), ranging from focal limb weak-ness to hemiplegia, followed by an altered level of conscious-ness in 33 patients (26 %), headache in 23 (18 %), dysphasiain 15 (12 %), gait disturbance in 14 (11 %) and seizures in 5(4 %). The majority of patients presented with a combinationof these symptoms. In the group of patients with ‘altered levelof consciousness,’ 7 were comatose (Markwalder grade 4).

The CSDH was left hemispheric in 62 patients (49 %),right hemispheric in 38 patients (30 %) and bilateral in 26patients (21 %). The CT density was mixed in 87 patients(69 %). In 16 patients it was homogeneously hypodense

Fig. 2 Age and gender distribution of the patient population

Table 1 Characteristics of different outcome groups

Characteristics Recurrence Complication Mortal

Mean age 75.0 74.5 77.5

Number of patients 11 (8.7 %) 43 (34.1 %) 17 (13.5 %)

Male/female 9/2 27/14 12/5

Anticoagulant 2 (18.2 %) 9 (20.9 %) 4 (23.5 %)

Antiplatelet 3 (27.3 %) 17 (39.5 %) 9 (52.9 %)

Coagulopathy 0 1 (2.3 %) 3 (17.6 %)

Ethylism 1 (9.1 %) 8 (18.6 %) 3 (17.6 %)

AHT 8 (72.7 %) 28 (65.1 %) 13 (76.5 %)

DM 4 (36.4 %) 7 (16.3 %) 6 (35.3 %)

HF 2 (18.2 %) 10 (23.3 %) 7 (41.2 %)

RF 1 (9.1 %) 7 (16.3 %) 5 (29.4 %)

COPD 1 (9.1 %) 6 (14.0 %) 2 (11.8 %)

CVA 2 (18.2 %) 8 (18.6 %) 6 (35.3 %)

ND 1 (9.1 %) 11 (25.6 %) 7 (41.2 %)

Dyslipidemia 3 (27.3 %) 8 (18.6 %) 5 (29.4 %)

CT hypo 2 (18.2 %) 10 (24.4 %) 2 (11.8 %)

Iso 4 (36.4 %) 10 (24.4 %) 3 (17.6 %)

Mixed 5 (45.5 %) 23 (56.1 %) 12 (70.6 %)

Bilateral 2 (18.2 %) 13 (30.2 %) 2 (11.8 %)

Left 7 (63.6 %) 13 (30.2 %) 7 (41.2 %)

Right 2 (18.2 %) 17 (39.5 %) 8 (47.1 %)

Acta Neurochir (2014) 156:981–987 983

(13 %) and homogeneously isodense in 23 patients (18 %).The differences in characteristics of patients needing repeatsurgery, having complications or in whom there was a fataloutcome are listed in Table 1.

Outcome

Overall 106 out of 126 patients (84 %) showed clinical im-provement at discharge, as shown by a decrease in theMarkwalder score postoperatively; 2 % had a Markwalderscore that did not alter after therapy. Fourteen percent had aworse outcome, which meant a deadly outcome in all but onepatient. At discharge, 72 patients (57 %) were asymptomatic(Markwalder 0), and 29 (23 %) had minimal residual symp-toms (Markwalder 1). Five patients (4 %) had a neurologicaldeficit (Markwalder 2), and three patients (2 %) had severalfocal signs (Markwalder 3) (Table 2). Mean hospital stay was15,4±12.5 days.

Recurrence of the subdural hematoma requiring surgeryoccurred in 11 patients (8.7 %).

Univariate analysis could not reveal a significant correla-tion between recurrence and one of the comorbidity factors,risk factors or CT characteristics of the subdural hematoma.

Postoperative complications occurred in 43 patients(34,1 %). They consisted of 27.8 % (35/43) of the medicalcomplications; 6.3 % (8/43) were surgical complications. Inthe 35 patients who did a medical complication, 69% sufferedpulmonary infections and 25 % urinary tract infections; theother 6 % had a decubitus ulcer or deep venous thrombosis.

The surgical complications included two wound infections,four epileptic seizures, an intracerebral hemorrhage and atension pneumocephalus (Table 3).

The only significant correlation was found with the pres-ence of a neurodegenerative disease (p=0,018).

A logistic regression analysis of the different Markwaldergrades showed that a higher preoperative Markwalder gradecorrelated with a higher risk of complications (Table 4).

Mean follow-up after discharge was 123.5 weeks (medianfollow-up 73 weeks; range 2.6-440 weeks). Two (1.8 %) outof 109 surviving patients were lost to follow-up, meaning theywere not seen at the outpatient clinic after discharge.

Overall 17 patients died (13.5 %) during their hospital stay:11 patients because of a pulmonary infection (8.7 %). Twodeaths were directly related to the surgical management: onedeveloped an intracerebral hemorrhage, and another patientdeveloped a tension pneumocephalus. Four others were al-ready comatose and did not wake up after surgery.

Fig. 3 Symptoms and signs atpresentation

Table 2 Markwalder score preoperatively and at discharge

PreoperativeMarkwalder

Markwalder atdischarge

N % N %

Grade 0 0 0 72 57

Grade 1 31 26 29 23

Grade 2 74 62 5 4

Grade 3 14 6 3 2

Grade 4 7 5 0 0

Death 17 13

Better 106 84

Constant 2 2

Worse (including death) 18 14

Table 3 Medical and surgical related complications of mini-craniotomy

Medical N (%) Mortality (%)

Pulmonary infection 24 (19.0) 11 (8.7)

Urinary tract infection 9 (7.1) 0 (0)

Decubitus wound 1 (0.8) 0 (0)

Deep venous thrombosis 1 (0.8) 0 (0)

Surgical N

Wound infection 2 (1.6) 0 (0)

Seizures 4 (3.2) 0 (0)

Pneumocephalus 1 (0.8) 1 (0.8)

Intracerebral hemorraghe 1 (0.8) 1 (0.8)

Total 43 (34.1) 13 (10.3 %)

984 Acta Neurochir (2014) 156:981–987

The presence of arterial hypertension (p=0,038), heartfailure (p=0,02), renal failure (p=0,017), neurodegenerativedisease (p=0,001), cerebrovascular accident (p=0,008) andcoagulopathy (p=0,019) seemed significantly related to mor-tality in univariate analysis. However, multivariate analysiscould not confirm a significant correlation between thesecomorbidities and mortality. Recurrence was not associatedwith a higher mortality.

Discussion

After the landmark paper by Svien and Gelety in 1964 thatconcluded that BHC is as effective as craniotomy but withlower mortality and morbidity, the craniotomy technique waslargely abandoned, and BHC became the first surgical treat-ment of choice [27]. It should be noted however that in theirtrial only 19 patients were treated by a large craniotomy withtotal membranectomy, which is a significantly different tech-nique frommini-craniotomy with membranotomy. This groupwas then compared to a group of 50 patients treated withBHC, indicating that selection bias plays a significant role inthis retrospective analysis.

Because of the more widespread use of CT and MRInowadays, CSDH is more frequently diagnosed, and the di-agnosis is more often made at an early stage. The presentpatient population is probably in a better general conditionthan those included in the study of Svien and Gelety. Inaddition, better neuro-anesthesia and progress in intensivecare have vastly decreased both operative and postoperativerisks for the neurologically critically ill patient [9]. Therefore,we believe the conclusions of that paper, published in 1964,cannot be translated to the current management of CSDH.

Recent meta-analyses have attempted to rationalize theoptimal treatment, but good quality data from well-designedstudies are still missing [7, 14, 30]. Lega et al. constructed adecision analysis model, trying to overcome the paucity ofquality data and heterogeneity of the published results [14].They concluded that BH was the most efficient choice forsurgical drainage of uncomplicated CSDH. They did mentionhowever that craniotomy has fewer recurrences but results insignificantly more frequent and serious complications. It bears

noting that the studies used to evaluate the craniotomy tech-nique rarely make a distinction between mini-craniotomy,large craniotomy, craniectomy or even burr hole treatment,making a correct evaluation of the mini-craniotomy techniqueimpossible [7, 14, 30]. In addition, craniotomy is mostlyreserved for recurrent CSDHs or if membranes or large acuteblood are visualized on preoperative imaging, leading to animportant selection bias. Weigel et al. confirmed that thestudies used to evaluate the craniotomy technique may leadto a distorted picture of the recurrence, mortality and morbid-ity [30]. To our knowledge, this is the first study where mini-craniotomy was performed in a consecutive series of all typesof CSDH.

In the present study, the mean age of the patients sufferingfrom CSDH was 73.4±14.2 years. There was a male predom-inance of 67.5%. A therapy with anticoagulants or antiplateletagents was present in 44 % (n=56), chronic alcohol abuse in14 % (n=18) and a coagulopathy in 5.5 % (n=7). In 75patients (59%) a history of trauma could be elicited; 4 patientshad a ventriculo-peritoneal shunt as the leading cause. Thesedata are consistent with results from previously publishedstudies [23].

Recurrence is a common problem in the treatment ofCSDH. Earlier publications mention recurrence rates rangingfrom 0.35 % to 60 %. In their evidence-based review of 48publications describing treatment of CSDH, Weigel et al.calculated a composite recurrence rate of 14.6 % for alltreatment modalities combined.

We have a significantly lower recurrence rate of only8.7 %. In our view, this is a result of the better visualizationof the subdural space and the consequential possibility to openmore septae and constricting membranes and to cauterizebleeding vessels.

Intuitively onemight think that anticoagulant or antiplatelettherapy might increase the recurrence rate, but studies inves-tigating this topic have not been able to confirm this [3, 15]. Inour series, we could not establish any relationship between theuse of these anti-clotting agents and recurrence either.

Predictive factors for recurrence have been extensivelyreviewed, encompassing pre-, peri – and postoperative fea-tures [25]. A recent prospective study assessed the relationshipof CT scan features and recurrence of CSDH. They found thatpreoperative hematoma volume and certain types (isodense,hyperdense, laminar and separated) had a higher risk of recur-rence [25]. Perioperative findings show that the presence of athick membrane visualized during surgery is also predictive[20]. These findings are not surprising since they reflect thestage of maturation of the hematoma and the risk of rebleeding[19]. Since a craniotomy offers better results in CSDH with ahigher risk for recurrence, it has been suggested to perform apreoperative MRI in order to better visualize theintrahematomal structure and adapt the neurosurgical tech-nique for every type of hematoma [28]. We did not find a

Table 4 Logistic regression table of Markwalder grade preoperativelyand complications postoperatively

Markwalder Odds ratio p-value Lower 95 % CI Upper 95 % CI

Grade 2 versusgrade 1

1.879 0.225 0.679 5.200

Grade 3 versusgrade 1

5.556 0.015 1.393 22.162

Grade 4 versusgrade 1

25.00 0.006 2.514 248.575

Acta Neurochir (2014) 156:981–987 985

significant relation between CTcharacteristics and recurrence,but since only 11 patients needed reoperation, these resultsshould be interpreted with caution.

Our low recurrence rate could on the other handindicate that mini-craniotomy is a suitable techniquefor all types of CSDH.

The possibility exists that some patients presented with arecurrence at a different hospital, but with only two patientslost to follow-up and two patients with a follow-up period lessthan 30 days, this would not change the outcome significantly.We also have a substantially longer follow-up period thanmost other studies.

Many publications have focused on the efficacy of drain-age of minimally invasive procedures, but rarely discusscomplications. Rohde et al. did focus on the complicationrates. In their retrospective study, they analyzed 376patients operated on with BHC, describing surgicalcomplications in 77 patients (20.5 %) and medical com-plications in 59 patients (15.7 %), giving a total com-plication rate of 136 patients (36.1 %) [21]. The overallrate is in line with our complication rate (34.1 %);however, we mainly have medical complications(27.8 %) and few surgical complications (6.3 %). Itbears noting that of all medical complications, 15.2 %were easily treated urinary tract infections. (Table 3).

Although rarely discussed, complications are of great clin-ical relevance as they are the leading cause of mortality. In ourseries, 11 out of 17 patients (65 %) died because of a pulmo-nary infection, and 2 patients died from their surgical compli-cations. One might assume that a less invasive techniquewithout the need for general anesthesia could lower the com-plication rate. This idea has been challenged by a studycomparing the bedside twist drill with a combinedgroup of BHC and craniotomy; complication rates wereshown to be similar between the two groups (36.3 inthe TDC group versus 33.3 in the combined BHC andcraniotomy group) [10].

As CSDH might be a “marker” of underlying chronic dis-ease, a “sentinel event” exposing decreased function of impor-tant organ systems, it is not excluded that some of these medicalcomplications were already present before surgery [17].

The presence of dyslipidemia did seem to have a protectiveeffect with regards to complications; further investigationshowed that these patients took statins. This finding shouldbe interpreted with caution since only 37 patients had dyslip-idemia, but a pleiotropic effect is not excluded. Patients underanticoagulant or antiplatelet treatment did not suffer a higherrate of complications. This finding is consistent with theliterature [4].

Our data show that the preoperative neurologic status is animportant prognostic risk factor for the development of com-plications. This is probably due to the longer hospitalizationstay of the critically ill patient.

An overall mortality rate of 13.5 % was found. Univariateanalysis of the comorbidity factors and mortality showeddifferent significant relationships [arterial hypertension(p=0.038), heart failure (p=0.02), renal failure (p=0.017],neurodegenerative disease (p=0.001), cerebrovascular acci-dent (p=0.008) and coagulopathy (p=0.019). Nonetheless,multivariate analysis was not able to confirm any significantrelation, presumably because of the fact that these older pa-tients have multiple comorbidity factors that are related toeach other. Another explanation may be the relatively lownumber of patients in each category.

In a study by Ramachandran et al., age, GCS at presenta-tion and associated illnesses such as cardiac and renal failurewere significantly related to mortality. Our own results doindeed confirm a significant relationship between mortalityand heart failure (p=0.02) and renal failure (p=0.017). Sur-prisingly, we did not see a linear relation between theMarkwalder grade and mortality.

Despite the mortality rate of 13.5 %, 84 % of the patientswere successfully treated and discharged home with 57 %having no symptoms (Markwalder 0) and 23 % having onlymild neurological symptoms (Markwalder 1).

As results of recent literature show a significant variation inreported mortality rates (3 %–24 %), it is difficult to saywhether this mortality rate is high or low [5, 6, 18, 26].

Mini-craniotomy has already been proposed as an effectivecompromise between the ‘minimally invasive’ burr hole and‘aggressive’ large craniotomy [1]. In our study, no relation-ship could be found between hematoma characteristicson CT and higher recurrence, morbidity or mortality inthis study, suggesting that mini-craniotomy is indeed asuitable technique for any type of CSDH, whetherisodense, hyperdense, hypodense or mixed with mem-branes. For the above-mentioned reasons, it is alreadysuggested as the primary technique for the treatment ofCSDH in children [13].

The present study is retrospective in nature, making itpotentially subject to sources of bias. As mini-craniotomywas exclusively used to treat all types of CSDH in allconsecutive patients, selection bias was avoided. Anoth-er strength is the comparatively large sample size of126 evaluable patients. The limitations of a retrospectiveseries still stand, however, and in order to determinewhether one technique is superior to the other, a wellset up prospective randomized controlled trial compar-ing mini-craniotomy with burr hole treatment and twistdrill craniostomy should be performed.

Conclusion

This is the first published series of CSDH in which all con-secutive patients were operated on by mini-craniotomy.

986 Acta Neurochir (2014) 156:981–987

Our retrospective analysis indicates that mini-craniotomy is an effective surgical technique for alltypes of CSDH, with results that compare favorably toother, more frequently used intervention types. It shouldtherefore not be neglected as a primary surgical tech-nique in the treatment of CSDH.

An RCT comparing different techniques for every type ofCSDH is indicated in order to identify the best surgical strat-egy in the treatment of CSDH.

Competing interests The authors have declared that no competinginterests exist.

References

1. Al-Jehani H, Petrecca K (2013) The case for mini-craniotomy in themanagement of chronic subdural hematoma. Acta Neurochir (Wien)155(1):189–190

2. Asghar M, Adhiyaman V, Greenway MW, Bhowmick BK, Bates A(2002) Chronic subdural haematoma in the elderly—a North Walesexperience. J R Soc Med 95(6):290–292

3. Aspegren OP, Astrand R, Lundgren MI, Romner B (2013)Anticoagulation therapy a risk factor for the development of chronicsubdural hematoma. Clin Neurol Neurosurg 115(7):981–984

4. Baraniskin A, Steffens C, Harders A, Schmiegel W, Schroers R,Spangenberg P (2013) Impact of pre-hospital antithrombotic medi-cation on the outcome of chronic and acute subdural hematoma. JNeurol Surg A Cent Eur Neurosurg 132(2):E48–E57

5. Callovini GM, Bolognini A, Callovini G, Gammone V (2013)Primary enlarged craniotomy in organized chronic subdural hemato-mas. Neurol Med Chir (Tokyo)

6. de Araújo Silva DO, Matis GK, Costa LF, Kitamura MA, deCarvalho Junior EV, de Moura Silva M, Barbosa BJ, PereiraCU, da Silva JC, Birbilis TA, de Azevedo Filho HR (2012)Chronic subdural hematomas and the elderly: surgical resultsfrom a series of 125 cases: old “horses” are not to be shot!Surg Neurol Int 3:150

7. Ducruet AF, Grobelny BT, Zacharia BE, Hickman ZL, Derosa PL,Anderson K, Sussman E, Carpenter A, Connolly ES (2012) Thesurgical management of chronic subdural hematoma. NeurosurgRev 35(2):155–169

8. Foelholm R, Waltimo O (1975) Epidemiology of chronic subduralhaematoma. Acta Neurochir (Wien) 32(3–4):247–250

9. Hamilton M, Frizzell J, Tranmer B (1993) Chronic subdural hema-toma: the role for craniotomy reevaluated. Neurosurgery 33(1):67–72

10. Horn EM, Feiz-Erfan I, Bristol RE, Spetzler RF, Harrington TR (2006)Bedside twist drill craniostomy for chronic subdural hematoma: acomparative study. Surg Neurol 65(2):150–153, discussion 153–4

11. Karibe H, Kameyama M, Kawase M, Hirano T, Kawaguchi T,Tominaga T (2011) Epidemiology of chronic subdural hematoma.No Shinkei Geka 39(12):1149–1153

12. Kim J-H, KangD-S, Kim J-H, KongM-H, SongK-Y (2011) Chronicsubdural hematoma treated by small or large craniotomy with mem-branectomy as the initial treatment. J Korean Neurosurg Soc 50(2):103–108

13. Klimo PJ, Matthews A, Lew SM, Zwienenberg-LeeM, Kaufman BA(2011) Minicraniotomy versus bur holes for evacuation of chronicsubdural collections in infants—a preliminary single-institution ex-perience. J Neurosurg Pediatr 8(5):423–429

14. Lega BC, Danish SF, Malhotra NR, Sonnad SS, Stein SC (2010)Choosing the best operation for chronic subdural hematoma: a deci-sion analysis. J Neurosurg 113(3):615–621

15. Lindvall P, Koskinen L-OD (2009) Anticoagulants and antiplateletagents and the risk of development and recurrence of chronic sub-dural haematomas. J Clin Neurosci 16(10):1287–1290

16. Markwalder TM, Steinsiepe KF, Rohner M, Reichenbach W,Markwalder H (1981) The course of chronic subdural hematomasafter burr-hole craniostomy and closed-system drainage. J Neurosurg55(3):390–396

17. Miranda LB, Braxton E, Hobbs J, Quigley MR (2011) Chronicsubdural hematoma in the elderly: not a benign disease. JNeurosurg 114(1):72–76

18. Mulligan P, Raore B, Liu S, Olson J (2013) Neurological and func-tional outcomes of subdural hematoma evacuation in patients over70 years of age. J Neurosci Rural Pract 4(3):250–256

19. Nakaguchi H, Tanishima T, Yoshimasu N (2001) Factors in thenatural history of chronic subdural hematomas that influence theirpostoperative recurrence. J Neurosurg 95(2):256–262

20. Ramachandran R, Hegde T (2007) Chronic subdural hematomas–causes of morbidity and mortality. Surg Neurol 67(4):367–372, dis-cussion 372–3

21. Rohde V, Graf G, Hassler W (2002) Complications of burr-hole craniostomy and closed-system drainage for chronic sub-dural hematomas: a retrospective analysis of 376 patients.Neurosurg Rev 25:89–94

22. Sambasivan M (1997) An overview of chronic subdural hematoma:experience with 2300 cases. Surg Neurol 47(5):418–422

23. Santarius T, Kirkpatrick PJ, Kolias AG, Hutchinson PJ (2010)Working toward rational and evidence-based treatment of chronicsubdural hematoma. Clin Neurosurg 57:112–122

24. Smely C, Van Velthoven V (1997) Comparative study of two cus-tomary cerebrospinal fluid shunting systems in early childhood hy-drocephalus. Acta Neurochir (Wien) 139(9):875–881, discussion 882

25. Stanišić M, Hald J, Rasmussen IA, Pripp AH, Ivanović J, Kolstad F,Sundseth J, Züchner M, Lindegaard K-F (2013) Volume and densitiesof chronic subdural haematoma obtained from CT imaging as predic-tors of postoperative recurrence: a prospective study of 107 operatedpatients. Acta Neurochir (Wien) 155(2):323–333, discussion 333

26. Stippler M, Ramirez P, Berti A, Macindoe C, Villalobos N, Murray-Krezan C (2013) Chronic subdural hematoma patients aged 90 yearsand older. Neurol Res 35(3):243–246

27. Svien H, Gelety J (1964) On the surgical management of encapsu-lated subdural hematoma. J Neurosurg 21:172–177

28. Tanikawa M, Mase M, Yamada K, Yamashita N, Matsumoto T,Banno T, Miyati T (2001) Surgical treatment of chronic subduralhematoma based on intrahematomal membrane structure on MRI.Acta Neurochir (Wien) 143:613–619

29. Weigel R, Krauss J, Schmiedek P (2004) Concepts of neurosurgicalmanagement of chronic subdural haematoma: historical perspectives.Br J Neurosurg 18(1):8–18

30. Weigel R, Schmiedek P, Krauss JK (2003) Outcome of contemporarysurgery for chronic subdural haematoma: evidence based review. JNeurol Neurosurg Psychiatry 74(7):937–943

31. Williams GR, BaskayaMK,Menendez J, Polin R,Willis B, Nanda A(2001) Burr-hole versus twist-drill drainage for the evacuation ofchronic subdural haematoma: a comparison of clinical results. JClin Neurosci 8(6):551–554

Acta Neurochir (2014) 156:981–987 987