To Trick or Treat: Antimicrobial Prophylaxis for Prevention of Meningitis in

Traumatic Skull Fractures Associated with CSF Leak

Hannah J. Davis, Pharm.D.

PGY1 Pharmacy Resident

University Health System, San Antonio, Texas

Division of Pharmacotherapy, The University of Texas at Austin College of Pharmacy

Pharmacotherapy Education and Research Center

University of Texas Health Sciences Center at San Antonio

October 31, 2014

Learning Objectives:

1. Recall anatomy and physiology of skull bones and intracranial components

2. Describe pathophysiology of skull fractures with associated cerebrospinal fluid (CSF) leak

3. Evaluate role of antibiotic prophylaxis in the setting of CSF leak

4. Indicate appropriate use of antibiotic prophylaxis with CSF leak

Davis 2

Background

I) Why do we care?

A) CNS infections including meningitis can be associated with cerebrospinal fluid (CSF) leaks

B) Overall incidence is low

C) Associated with morbidity, mortality, and increased cost

D) Controversial role for antibiotic prophylaxis

Figure 1. Severity of Injuries

II) Traumatic skull fractures

A) Incidence

i) 1.7 million traumatic brain injuries (TBI) annually1

(a) 275,000 hospitalizations for nonfatal TBI2

(b) 52,000 TBI-related deaths

(c) 80,000-90,000 experience long-term disability

(d) Mechanism of injury1

(1) Fall – 28%

(2) Motor vehicle accident – 20%

(3) Blunt force – 19%

(4) Assault – 11%

(5) Blast injuries

(e) Severity of TBI based on Glasgow Coma Scale (GCS)

(1) Presence of skull fractures is a prognostic factor for worse outcome3

(i) Greater force of impact with trauma

ii) 30% of patients with TBI have associated skull fractures

(a) Incidence of fracture by type

(1) Basal skull fractures (BSF) 3.5-25%4,5

(i) Incidence lower in pediatric patients

1. Increased flexibility of skull base

2. Underdevelopment of ethmoid, frontal, and mastoid air cells

(2) Maxillofacial fractures 23%5

(3) Cranial vault fractures 1%5

Traumatic brain injury

(TBI) Skull fractures CSF fistula

Central nervous

system (CNS) infections

Davis 3

B) Morbidity

i) Severe TBI with skull fractures is associated with a moderate or poor outcome6,7

(a) Intensive care unit admission, ventilator support during recovery, increased risk for

prolonged hospital length of stay8

ii) Permanent neurologic disorders6

(a) Epilepsy, blindness, deafness, diplopia, facial paralysis, other cranial nerve deficits9,10

iii) Traumatic encephalocele

(a) Herniation of brain tissue

iv) Pneumocephalus4,11

(a) Entrapment of air under pressure in intracranial cavity

v) Cerebrospinal fluid leaks

(a) Escape of fluid that surrounds the brain and spinal cord

vi) Infectious complications12

(a) Meningitis, encephalitis, intracranial abscesses, subdural empyema

(b) Head trauma is the most common cause of recurrent bacterial meningitis

C) Mortality

i) Early mortality has substantially declined over time7

ii) 7% mortality in mild TBI with BSF13

iii) 22% mortality in patients with penetrating injury and CSF leak14

iv) 65% mortality in severe TBI with skull fractures3

(a) 30-54% mortality in severe TBI

D) Cost5

i) Average cost of hospital stay in TBI cohorts from 2005-2009

(a) Without meningitis: $32,319

(b) With meningitis: $42,808

ii) Independent risk factors for escalating cost of hospitalization

(a) Rhinorrhea OR 2.0 (CI 1.6-2.7) and otorrhea OR 2.3 (1.9-2.7)

(b) Posttraumatic meningitis OR 3.1 (CI 2.5-3.8)

(c) Surgical intervention

E) Fracture classifications1

i) Characteristics

(a) Linear

(b) Depressed

(c) Compound

(d) Open

(e) Closed

(f) Penetrating

Davis 4

ii) BSF

(a) Characteristics

(1) Thin and more vulnerable to trauma

(2) Air-filled sinuses and multiple foramen render skull base more vulnerable to trauma

(b) Anterior BSF

(1) Type I, cribriform

(2) Type II, frontoethmoidal

(3) Type III, lateral frontal

(4) Type IV, complex

(c) Middle BSF

(1) Longitudinal - parallel to petrous portion of temporal bone

(2) Transverse - perpendicular to petrous portion of temporal bone

(3) Sella turcica fractures

(d) Posterior BSF

(1) Clivus

(i) Transverse, oblique, longitudinal

(2) Occipital

(i) Ring fracture- encircles foramen magnum at skull base

(ii) Occipital condyle- type I – IV

iii) Facial fractures15

(a) Le Fort I

(1) Horizontal maxillary fracture, separating the teeth from upper face, fracture line

passes through alveolar ridge, lateral nose and inferior wall of maxillary sinus

(b) Le Fort II

(1) Pyramidal fracture, passes through posterior alveolar ridge, lateral walls of maxillary

sinuses, inferior orbital rim and nasal bones

(c) Le Fort III

(1) Craniofacial disjunction, fracture line passes through nasofrontal suture, maxillo-

frontal suture, orbital wall and zygomatic arch

Figure 2. Le Fort Fracture Classifications

www.hakeem-sy.com/main/node/26408

Le Fort I Le Fort II Le Fort III

Davis 5

III) CSF

A) Function16

i) Provide physical support and buoyancy for brain/spinal elements

ii) Remove by-products of metabolism from neural tissues and regulates chemical

environment of brain

B) Production and circulation16,17

i) Produced by choroid plexus in the lateral, third, and fourth ventricles at a rate of 20-30

mL/h

(a) Approximately 90-150 mL total volume

ii) Flows from: lateral ventricles to third ventricle to fourth ventricle through subarachnoid

space and circulates through meninges before being reabsorbed via arachnoid granulations

IV) CSF leaks

A) CSF fistula5

i) A fracture involving skull or facial structures may produce a breach in dura

ii) Abnormal passageway allowing escape of CSF from subarachnoid space

iii) Can be associated with CSF leak

B) Types of CSF leaks

i) Rhinorrhea18,19

(a) 72% of CSF leaks

(b) CSF leak through nares

(c) Origins of leak19,20

(1) Fistula of anterior cranial fossa

(2) Fistula of temporal bone

(3) Fistula of middle cranial fossa through sphenoid sinus

(d) Presentation19,20

(1) Fluid initially may appear sanguineous

(2) Unilateral or bilateral clear fluid

(3) Discharge amount varies from drops to stream-like flow

(4) Reserviour sign

(i) Rush of fluid when patient brought from supine to upright with neck flexed

(ii) Useful for collection of diagnostic samples

(5) Anosmia19

(i) Inability to perceive odor

(ii) Present in 80% of cases or rhinorrhea and only 5% in TBIs

(6) Target sign or halo effect19,21

(i) CSF will migrate further when CSF/blood applied to filter paper

1. Apparent on pillowcase or sheets

(ii) Bull’s eye stain appearance

(7) Battles sign

(i) Periorbital ecchymosis

Davis 6

ii) Otorrhea

(a) 28% of CSF leaks18

(b) CSF leak through ear

(1) Typmpanic membrane may be intact or ruptured22

(i) Healing of tympanic membrane occurs within 1 week in ~80%23

(c) Origins of leak

(1) BSF involving temporal bone22

(d) Presentation21,23

(1) Clear fluid draining from ear

(2) Conductive hearing loss

(i) 20% still present 7 years post trauma

(3) Sensation of fluid or bubbles in ear

(4) Hemotympanum

(i) Presence of blood in tympanic cavity of middle ear

(5) Battles sign

(i) Retroauricular ecchymosis

C) Diagnosis of CSF leaks

i) High clinical suspicion of CSF leak in patients with skull fractures24

(a) Prompt recognition is key for instituting treatment and preventative measures9,21

(b) Potential to be overlooked due to nature of concomitant issues

ii) Diagnostic markers can confirm presence of CSF leak25

(a) Produced in CSF and when present in sample indicate presence of CSF

(1) Beta-2 transferrrin

(2) Beta-trace protein

iii) Confirm presence of concomitant fractures with imaging

(a) Computed tomography (CT) scan

(b) Formerly, X-ray

D) Incidence of CSF leak

i) 1-3% of all TBIs5,16,22

ii) 9 - 11% of penetrating head injuries4,20

iii) 10-30% of BSF6,25

iv) 36% of Le Fort fractures20

E) Time to presentation of CSF leak25

i) Acute

(a) Within 2 days following injury – 50%

ii) Delayed

(a) Within 1 week – 70%

(b) Within 3 months – 99%

iii) Sources of delayed presentation19,20

(a) Edema, inflammation, hematoma obstructing flow

(b) May appear in times of increased intracranial pressure (ICP)

Davis 7

F) Duration of CSF leak

i) Temporary

(a) Less than 24 hours

ii) Persistent

(a) Present greater than 7 days

(b) Characteristics of injury which prevent arachnoid mesh from sealing to dura21,26,27

(1) Fracture site mobility

(2) Disrupted blood supply

(3) Infections

iii) Intermittent

(a) Soft tissue or brain can obstruct flow

G) Treatment of CSF leak with skull fractures

i) Reduction and fixation of facial fractures as appropriate20

(a) Stabilize fracture and prevents disruption of dura by reducing movement of facial

skeleton

(b) Facilitate body’s natural repair of anterior fossa CSF leak

ii) Conservative treatment16,20

(a) Facilitate spontaneous resolution of CSF leak

(1) Adhesion formation and granulation from local inflamed meninges

(2) Mucosal swelling, local fibrosis, regenerated nasal mucosa

(3) Dura mater dose not regenerate

(b) Strict CSF precautions20,25

(i) Bed rest

(ii) Avoid: nose blowing, Valsalva maneuvers, straws, incentive spirometer

1. To prevent increase intracranial pressure

(2) Head elevation16,25

(i) At least 15º

(ii) To prevent negative gradient allowing ascending path for bacteria between

intracranial and paranasal sinus cavities

(c) Duration of treatment is controversial

(1) Typically pursued for 1 week

(d) Spontaneous resolution outcomes

(1) Within 3 days in 40% of all cases22

(2) Within 1 week in 70% of rhinorrhea16,22,28

(3) Within 10 days in 85% of rhinorrhea16

(4) Ultimately in 98% of rhinorrhea

(5) Otorrhea ceases spontaneously in nearly all cases

(6) Spontaneous closure higher with temporal bone fractures than BSF

Davis 8

iii) CSF diversion

(a) If conservative treatment fails25

(i) Typically after 1 week29

(b) Treatment options

(1) Temporary CSF drainage device placed

(i) Lumbar drain

(ii) External ventricular drain (EVD)

(iii) Duration less than 7 days due to increased infection risk

iv) Surgical management4,16

(a) Employed for definitive closure of CSF leak

(1) Failure of conservative management and CSF diversion

(b) Approaches

(1) Advances in neurosurgical approaches from open craniotomies to endoscopic

procedures

(2) Surgical approach and graft type used to repair leak depends on individual clinical

scenario

(i) Based on force of impact, location of fistula, temporal relationship between the

leak and traumatic event

(ii) Endoscopic endonasal approach preferred

(iii) Transcranial approach, extracranial approach, suprasinus transfrontal approach

also available

(c) Outcomes25

(1) 75-100% success rate

V) CNS Infections4,20,22,30

A) CSF fistulas allow direct access of nasophyaryngeal and middle ear flora to the subarachnoid

space

B) Pathogens30,31

i) Most common: Streptococcus pneumoniae, Haemophilus influenzae

ii) Less common: gram-negative, more resistant pathogens

C) Can appear within hours or up to years after injury4,19,24

i) Median onset 5-13 day

D) Risk factors for meningitis

i) CSF leaks

(a) Not associated with severity of leak or size of dural tear19,20

(b) Rhinorrhea increases risk 23-fold5

(c) Otorrhea increases risk 9-fold

(d) Persistent leaks increase risk 9-fold16,24,28,32,33

(1) Improper healing or traumatic encephaloceal

ii) Pneumocephalus increase risk up to 4-fold11,16

Davis 9

iii) Fracture types

(a) Penetrating fractures14

(1) Compound fractures with skull depressed deeper than thickness of cranium12

(2) Incidence of infection ~ 10%34

(b) Frontobasilar fractures proximal to midline33

(c) Fracture displacement > 1 cm

iv) GCS < 829

(a) Associated with severity of injury and increased utilization of invasive interventions

v) Surgical interventions are associated with a 5-fold increase risk retrospectively5

(a) Craniotomies12

(1) Incidence 0.8-1.5%

(i) Risk associated with concomitant infection at site of incision and duration of

surgery greater than 4 hours

(b) EVDs

(1) Incidence 4-17%

(i) Increased risk associated with colonization of catheter at time of surgery

Figure 3. Incidence of Meningitis Associated with CSF Leaks in Literature10,17,22,24, 26,27,30-

43,46,49,50

52.4% (11 / 21)

19% (16 / 84)

3% (1 / 33)

22.2% (12 / 54)

8.7% (13 / 149)

4.7% (2 / 43)

1.9% (1 / 52)

3.4% (2 / 58)

0% (0 / 7)

6.1 % (2 / 33)

8.3% (3 / 36)

1.4% (2 / 138)

4% (2 / 50)

3.6% (1 / 28)

9.8% (21 / 215)

4% (2 / 50)

6.8% (6 / 88)

10.3% (10 / 97)

0% (0 / 12)

0% (0 / 34)

18.5% (15 / 81)

0% (0 / 9)

4.3% (2 / 47)

0% 10% 20% 30% 40% 50% 60% 70%

1942

1954

1966

1966

1973

1975

1976

1976

1978

1983

1986

1988

1991

1992

1993

1995

1995

1996

2004

2004

2004

2009

2010

Davis 10

E) Incidence

i) 0.2- 17.8% of TBI 5,11

ii) When CSF leak is present5,25

(a) Range: 0-52%

(b) Median: 4.24%,

(c) Rhinorrhea: 7-38%

(d) Otorrhea: 2.6%

iii) Penetrating injury with CSF leak: 36%22

(a) 1.7% without CSF leak

iv) 12- 50% of cases are associated with recurrence35

F) Mortality6,20

i) Early studies found morbidity and mortality rates of 50% and 25% respectively for

intracranial infections following CSF rhinorrhea

ii) Recent literature suggests posttraumatic meningitis is associated with ~10% mortality rate

VI) Role of antibiotic prophylaxis24

A) Early publications emphasized importance of antibiotic use as a means of reducing meningitis

i) Advocated for patients who presented with CSF leak or pneumocephalus

ii) Recommended duration of prophylaxis 5-7 days after cessation of CSF leak

iii) Recommendation was based on perceived or theoretical risk of infection and experience of

authors

B) Proposed mechanism of benefit48,49

i) Antibiotics maintain sterility of CSF until defect in dura closes spontaneously or is

surgically corrected

ii) Antibiotics may eradicate bacterial colonization in nasopharynx, nasal sinuses, and aural

canal

Literature Review

C) 24 studies evaluating incidence of meningitis in patients with CSF leak

i) 19 evaluate the role of antibiotic prophylaxis

ii) Literature dating back to 1942

Figure 4. Role of Antibiotics

16%

31%

53%

Position on Antibiotic Prophylaxis in Literature

Supportive

Neutral stance

Does not support

Davis 11

D) Conflicting meta-analysis published18,49

i) Brodie et al. 1997

(a) Objective: determine efficacy of antibiotic prophylaxis in CSF leaks

(b) Population: 6 studies involving 324 patients

(1) Included all articles with case series and sufficient data to allow evaluation

(2) Findings

(i) 237 patients received prophylactic antibiotics and 87 did not

(ii) Overall incidence of meningitis: 15/234 (6.4%)

1. Antibiotic prophylaxis: 6/237 (2.5%)

2. Inadequate/no antibiotic prophylaxis: 9/87 (10%)

3. Associated with p value=0.006

(3) Conclusion: meningitis rate was significantly lower in patients who received

antibiotic prophylaxis

ii) Villalobos et al. 1998

(a) Objective: determine efficacy of antibiotic prophylaxis in patients with BSF

(1) Evaluated CSF leaks as a secondary endpoint

(b) Population: 12 studies involving 1241 patients

(1) CSF leak: 9 studies involving 547 patients

(c) Findings in patients with CSF leaks

(1) 297 received prophylactic antibiotics and 250 did not

(i) 6 studies with 179 patients with either rhinorrhea or otorrhea were evaluated

(ii) Overall incidence of meningitis 63/547 (11.5%)

1. Antibiotic prophylaxis: 29/297 (9.7%)

2. Inadequate/no antibiotics: 34/250 (13.6%)

3. Common OR 1.34 (CI 0.75-2.41)

(d) Concluded antibiotic prophylaxis after BSF does not appear to decrease risk of

meningitis

VII) Primary Literature

Leech PJ et al. The Lancet. 1973;1:1013-101624

Objective Investigate incidence of meningitis, morbidity and mortality with adequate and inadequate

antibiotic prophylaxis

Trial design Retrospective, single center review at teaching hospital in Glasgow

Patients Inclusion: skull fracture with associated CSF leak from 1943-1972

Exclusion: not specified

Outcomes Incidence of meningitis, morbidity and mortality

Methods

Adequate prophylaxis: penicillin 500 mg PO QID and sulfadimidine 500 mg PO Q6H from time

of injury until 1 week after CSF leak cessation

Inadequate prophylaxis: those not meeting above criteria

Diagnostic criteria: not specified for skull fractures, CSF leak, or meningitis

Follow up: clinic visits, postal query, reviewing cases of meningitis at hospital; for up to ten years

Davis 12

Statistics Not specified

Results

n=155

Referred for neurosurgical care: 141/155

CSF leak associated with fracture: 108/155

Surgical repair of rhinorrhea: 65/118

Le Fort fractures: all received antibiotic prophylaxis, incidence not specified

* p ≤0.05

Increased incidence of meningitis associated with otorrhea if fracture was associated with

middle fossa (p=0.01)

All patients with meningitis recovered satisfactorily

Adequate

Prophylaxis

n=95

No Adequate

Prophylaxis

n=54

Rhinorrhea Number 76 42

Meningitis

leak < 1 week

0 (0%)

7 (17%)*

Otorrhea Number 19 12

Meningitis 1 (5%) 5 (41.7%) *

Mortality Overall 2/155 (3%)

Attributable 0/15 (0%)

Morbidity Anosmia Majority

Recurrent rhinorrhea 4/124

Pathogens Streptococcus pneumonia n=5, coliforms n=1, no growth n= 5

Authors’

conclusions

The inference is clear: all cases of CSF otorrhea should receive adequate antibiotic

prophylaxis from time of injury until 1 week after the leak has ceased.

The benefit of adequate chemotherapy in protecting patients from meningitis in patients with

rhinorrhea has been demonstrated.

Critique

Strengths

Large patient population

Majority of fractures associated with middle or anterior fossa

Morbidity and mortality were examined as endpoints

Limitations

Temporal bias: variations in surgical procedures, CSF precautions, antibiotic selection,

prevalence of resistant organisms

Included pre-existing cases of meningitis

High incidence of meningitis with otorrhea

Unknown exposure to antibiotics in patients receiving inadequate prophylaxis

Incidence of meningitis in rhinorrhea arm no longer associated with adequate or inadequate

prophylaxis following first week

Extensive referral from outside hospitals

Unknown baseline characteristics

Statistical analysis used not specified

Lack of specification of diagnostic techniques for fractures, CSF leak, meningitis

Take home

points

Results favored use of antibiotic prophylaxis

Questionable validity

Broad spectrum of antibiotics and extended duration of treatment

Davis 13

Ignelzi RJ et al. J Neurosurg. 1975;43:721-72650

Objective Primary: determine efficacy of antibiotic prophylaxis in treatment of BSF

Post hoc: determine effect of prophylactic antibiotics on nasopharyngeal flora

Trial design

Combined retrospective and observational prospective at Denver General Hospital

Retrospective arm: patients admitted with BSF, initiated on antibiotic prophylaxis (ampicillin,

cephalothin, or other)

Prospective arm: withheld antibiotic prophylaxis for BSF; antibiotic treatment was indicated if

fever, neck stiffness, clinical deterioration

Patients Retrospective arm: Adult patients with traumatic BSF who survived > 10 days

Prospective arm: Adult patients with BSF admitted for 10 day observation, excluded patients who

received antibiotics for other indications

Outcomes Rates of CNS and other other infections, nasopharyngeal cultures on admission, day 5 and 10

Methods

First year: Antibiotic prophylaxis (ampicillin or cephalothin 1 g Q6H IV x 3 days or until 2 days

after cessation of CSF leak then IM or PO antibiotics for duration of hospitalization

Posterior nasopharynx cultures: Post hoc endpoint, randomized to receive either no antibiotic,

ampicillin or cephalothin; obtained at hospital admission, day 5 and day 10

Follow up: method unspecified

Statistics

ANOVA

Fisher’s exact test

Chi-square

Results

n=129 with 136 BSF

n=40 with CSF leak

Follow up: 3-24 months

Baseline characteristics

Treated group had increased incidence facial fractures and one compound depressed fracture

No antibiotic prophylaxis group had more patients unconscious for > 24 hours, CNS

operations, prior medical illness

Prophylaxis: ampicillin: n=41 cephalothin: n=9 other: n=4

Prophylaxis

(Retrospective)

CSF leak n=17

No Prophylaxis

(Prospective)

CSF leak n=26

CNS infections 2 (12%) 0 (0%)

Pathogens isolated Resistant E. coli n=1,

no growth n=1

N/A

Attributable mortality 1 (50%) N/A

Attributable morbidity 1 (50%) N/A

Mortality, overall 7.2% 7.4%

Davis 14

Nasopharyngeal Cultures Results

Antibiotic Exposure Day 0 Day 5 Day 10

None (n=5) Normal Unchanged Unchanged

Cephalothin (n=3) Normal Gram-negative (n=2)

Increased normal (n=1)

Gram-negative (n=2)

Increased normal (n=1)

Ampicillin (n=2) Normal Gram-negative Gram-negative

Normal- Streptococcus and Staphylococcus sp.

Gram negative- resistant to antibiotic therapy

Outcomes Associated with Antibiotic Administration

Authors’

conclusions

Antibiotics are ineffective in preventing CNS infections and in some cases, harmful.

Antibiotics associated with more resistant, invasive and potentially pathogenic organisms in

nasopharynx.

Recommend close observation for early signs of meningitis and if present, start appropriate

antibiotics.

Critiques

Strengths

Included observational prospective arm

Post hoc investigation of antibiotic impact on nasopharyngeal flora

Limitations

Not specific to those with CSF leak, x-ray confirmation of fracture was present in 14/129

patients

Not adequately powered to determine no difference exists or more harm associated with

antibiotics, limited number of patients with CSF leak in 2 year time frame

Unknown demographic information; epidemiologic, fracture specific, time to onset/duration

of CSF, role of CSF precautions, duration of leak, duration of antibiotics or length of

hospitalization

More selective criteria for the prospective portion

Take home

points

No benefit demonstrated with antibiotic prophylaxis

Nasopharyngeal cultures demonstrated no sterilization of site

Change from usual nasopharyngeal flora to gram-negative/resistant bacteria

Alteration of potential pathogen

Alteration of flora to resistant bacteria

Nasopharyngeal flora not sterilized

Antibiotic administered

Davis 15

Eljamel MS. Br J Neurosurg. 1993;7:501-50535

Objective Evaluate efficacy of antibiotic prophylaxis in patients with CSF leak prior to surgical dural repair

Trial design Retrospective analysis of patients with CSF leaks who received adequate or inadequate antibiotic

prophylaxis prior to surgical dural repair at Mersey Regional

Patients Inclusion: adult patients with CSF fistula

Exclusion: patients who received antibiotics for reasons other than CSF leak, from time of surgical

dural repair forward

Outcomes Meningitis, mortality, and pathogens isolated

Methods

Similar age, CSF leak type, duration of leak, presence/absence of skull fractures

Adequate antibiotic prophylaxis: if started within 3 days of onset of CSF leak; penicillin 500 mg

Q6H and sulponamide 500 mg Q6H and continued for at least one week after CSF leak cessation

Meningitis diagnosis: based on clinical signs and symptoms, confirmed by isolating pathogen or

cytology of CSF

CSF leak: confirmed surgically or post-mortem

Follow-up: until surgical dural repair, death or 1990

Statistics

Chi-square test with Yate’s correction

Two-tailed p values

Log rank

Kaplan-Meier

Results

n= 253

Excluded= 38 patients for receiving antibiotics for other indications (pneumonia, facial

fixation) or were receiving antibiotics other than penicillin and sulponamide

Baseline characteristics

Mean age 35 years old, majority male, with rhinorrhea and skull fractures

More patients in the prophylaxis arm had facial fractures and pneumocephalus

Majority of patients received antibiotic prophylaxis within 24 hours of injury

Mean follow-up 2.5 years

Adequate Prophylaxis

n=106

Inadequate Prophylaxis

n=109

Facial fractures 52 26

Skull fractures 76 75

Rhinorrhea 99 97

Duration CSF leak, mean 17 days 20 days

Pneumocephalus 31 (29%) 12 (11%)*

Meningitis, first week 7 (6%) 10 (9%)

Meningitis, per year 8 (8%) 13 (12%)

Meningitis, total 20 (19%) 36 (33%)

Streptococcus pneumoniae 6 22

Gram-negative 3 0*

No pathogen isolated 11 14

Survival free from meningitis,

1 month

86 (81%) 84 (77%)*

Mortality 20 (19%) 25 (22%)

*p value <0.05

Davis 16

Critiques

Strengths

Large population

Specific to CSF leaks

Fracture type specified

Primary endpoint examining efficacy of antimicrobial prophylaxis

Excluded patients receiving antibiotics for other indications

Limitations

More pneumocephalus and facial fractures in antibiotic prophylaxis arm

Mean duration of CSF leak 17-20 days

Broad antibiotic coverage and extended duration of prophylaxis

Unknown demographic information, mode of injury, incidence of penetrating or other fracture

characteristics, number of patients receiving CSF diversion

Authors’

conclusions

These data do not support routine use of antibiotic prophylaxis in patients with CSF leak

It is ethically justifiable to withhold antibiotic prophylaxis in patients with CSF fistula until a

prospective controlled double-blind trial has settled the question

Take home

points No statistically significant in incidence of meningitis in a well-designed retrospective review

VIII) Limitations in literature

A) Lack of power of individual studies

i) Infrequent incidence of CSF leaks

ii) Even lower incidence of post-traumatic meningitis

B) Confounding factors in studies

i) Retrospective

ii) Lack of randomization

C) Temporal bias

i) Treatment of CSF leak

(a) CSF precautions

(b) CSF diversions and surgical management

ii) Antibiotic availability

iii) Alteration in pathogens

IX) Conclusions

A) Overall the data supporting the use of antibiotic prophylaxis to prevent meningitis in post

traumatic CSF leaks is of poor quality

B) Convincing evidence to support antibiotic prophylaxis is lacking

C) Randomized, prospective, multicenter trials are needed to provide further evidence

D) Harm has been demonstrated with administration of prophylactic antibiotics

Davis 17

X) Recommendation

i) Antibiotic prophylaxis in skull fractures with CSF leak not recommended to prevent

meningitis

(a) Closed fractures of basilar skull or skull vault

ii) Concomitant conditions may benefit from antibiotic prophylaxis

(a) Le Fort facial fractures, open/penetrating fractures, compound/depressed skull fractures

(b) Potential benefit in high-risk subsets of patients with multiple risk factors cannot be fully

excluded based on current level of evidence

(1) Lack of randomization

(2) High risk patients typically received antibiotic prophylaxis

(c) Antibiotic prophylaxis when selected should cover pathogens most likely to colonize

patient’s nasopharynx

(d) Duration should be one week after CSF leak resolution

iii) Regardless of presence of antibiotic prophylaxis, patients should be monitored for signs

and symptoms associated with meningitis

(a) If suspected, empiric treatment of meningitis should be promptly initiated

(1) Obtain cultures

(2) Vancomycin + cefepime

Davis 18

References

1. Head trauma. In: Advanced Trauma Life Support. 9th ed. American College of Surgeons. 2012;148-174.

2. Corrigan J, Selassie A, Orman J. The epidemiology of traumatic brain injury. J Head Trauma Rehabil. 2010; 25:

72-80.

3. Tseng W, Shih H, Su Y, et al. The association between skull bone fractures and outcomes in patients with severe

traumatic brain injury. The Journal of Trauma. 2011; 71: 1611-1614.

4. Tasdemiroglu E, Patchell R. Classification and management of skull base fractures. Neurosurgery Quarterly. 2002;

12: 42-62.

5. Sonig A, Thakur J, Chittiboina P, et al. Is posttraumatic cerebrospinal fluid fistula a predictor of posttraumatic

meningitis? A US nationwide inpatient sample database study. Neurosurg Focus. 2012; 32: 1-7

6. Perheentupa U, Kinnunen I, Grenman R, et al. Management and outcome of pediatric skull base fractures.

Otorhinolaryngology. 2010; 74: 1245-1250.

7. Maas A, Stocchetti N, Bullock R. Moderate and severe traumatic brain injury in adults. Lancet. 2008; 7: 728-741.

8. Helling T, Evans L, Fowler D. Infectious complications in patients with severe head injury. The Journal of Trauma.

1988; 28: 1575-1577.

9. Katzen J, Jarrahy R, Eby J, et al. Craniofacial and skull base trauma. The Journal of Trauma Injury, Infection, and

Critical Care. 2003; 54: 1026-1034.

10. Fishman G, Fliss D, Benjamin S, et al. Multidisciplinary surgical approach for cerebrospinal fluid leak in children

with complex head trauma. Childs Nerv Syst. 2009; 25: 915-923.

11. Eftekhar B, Ghodsi M, Nejat F, et al. Prophylactic administration of ceftriaxone for the prevention of meningitis

after traumatic penumocephalus: results of a clinical trial. J Neurosurg. 2004; 101: 757-761.

12. Van de Beek D, Brouwer C, Thwaites G, et al. Advances in treatment of bacterial meningitis. 2012; 380: 1693-

1702.

13. Koonsman M, Dunn E, Hughes K, et al. How much monitoring is needed for basilar skull fractures? The American

Journal of Surgery. 1992; 164: 487-490.

14. Management of cerebrospinal fluid leaks. J Trauma. 2001; 51: 29-33.

15. Noffze M, Tubbs S. Rene Le Fort 1896-1951. 2011; 24: 278-281.

16. Prosser J, Vender J, Solares C. Traumatic cerebrospinal fluid leaks. Otolaryngol Clin N Am. 2011; 44: 857-873.

17. Schlosser R, Bolger W. Nasal cerebrospinal fluid leaks. The Journal of Otolaryngology. 2002; 31: 28-36.

18. Brodie H. Prophylactic antibiotics for posttraumatic cerebrospinal fluid fistulae: a meta-analysis. Arch Otolaryngol

Head Neck Surg.1997; 123: 749-752.

19. Lewin W. Cerebrospinal fluid rhinorrhea in closed head injury. The British Journal of Surgery. 1954; 171: 1-18.

20. Marentette L, Valentino J. Traumatic anterior fossa cerebrospinal fluid fistulae and craniofacial considerations.

Current Issues in Head and Neck Trauma. 1991; 24: 151-163.

21. Chandler J. Traumatic cerebrospinal fluid leakage. Otolaryngologic Clinics of North America. 1983; 16: 623-631.

22. Davies M, Teo C. Management of traumatic cerebrospinal fluid fistula. The Journal of Cranio-Maxillofacial

Trauma. 1995; 1: 9-17.

23. Tos M. Course of and sequelae to 248 petrosal fractures. Acta Otolaryng. 1973; 75: 353-354.

24. Leech P, Paterson A. Conservative and operative management for cerebrospinal-fluid leakage after closed head

injury. The Lancet. 1973; 1: 1014-1015.

25. Ziu M, Savage J, Jimenez D. Diagnosis and treatment of cerebrospinal fluid rhinorrhea following accidental

traumatic anterior skull base fractures. Neurosurg Focus. 2012; 32: 1-17.

26. Choi D, Spann R. Traumatic cerebrospinal fluid leakage: risk factors and the use of prophylactic antibiotics. British

Journal of Neurosurgery. 1996; 10: 571-575.

27. Marentette L, Valentino J. Traumatic anterior fossa cerebrospinal fluid fistulae and craniofacial considerations.

Current Issues in Head and Neck Trauma. 1991; 24: 151-163.

28. Bell R, Dierks E, Homer L, et al. Management of cerebrospinal fluid leak associated with craniomaxillofacial

trauma. J Oral Maxillofac Surg. 2004; 62: 676-684.

Davis 19

29. Yilmazlar S, Arslan E, Kocaeli H, et al. Cerebrospinal fluid leakage complicating skull base fractures: analysis of

81 cases. Neurosurg. 2006; 29: 64-71.

30. Matschke J, Tsokos M. Post-traumatic meningitis: histomorphological findings, postmortem microbiology and

forensic implications. 2001; 115: 199-205

31. Ioannis B, Soultana T, Pavlos S. Posttraumatic meningitis: bacteriology, hydrocephalus, and outcome.

Neurosurgery. 1994; 35: 422-427.

32. Mincy E. Posttraumatic cerebrospinal fluid fistula of the frontal fossa. The Journal of Trauma. 1966; 6: 618-623.

33. Sakas D, Beale D, Ameen A, et al. Compound anterior cranial bas fractures: classification using computerized

tomography scanning as a base for selection of patients for dural repair. J Neurosurg. 1998; 88: 471-477.

34. Dunn L, Foy P. Anticonvulsant and antibiotic prophylaxis in head injury. Ann R Coll Surg Engl. 1994; 76: 147-

149.

35. Eljamel M. Antibiotic prophylaxis in unrepaired CSF fistulae. British Journal of Neurosurgery. 1993; 7: 501-506.

36. Adeleye A, Basilar skull fracture: outcome of acute care without antibiotic prophylaxis in a Nigerian neurosurgical

unit. Turkish Neurosurgery. 2010; 20: 430-436.

37. Clemenza J, Kaltman S, Diamond D. Craniofacial trauma and cerebrospinal fluid leakage: a retrospective clinical

study. J Oral Maxillofac Surg. 1995; 53: 1004-1007.

38. McGuirt W, Stool S. Cerebrospinal fluid fistula: the identification and management in pediatric temporal bone

fractures. Laryngoscope. 1995; 105: 359-364.

39. Demetriades D, Charalambides D, Lakhoo M. Role of prophylactic antibiotics in open and basilar fractures of the

skull: a randomized study. Injury. 1992; 5: 377-380.

40. Ash, G. Antimicrobial prophylaxis for fractured base of skull in children. Brain Injury. 1992; 6: 521-527.

41. Frazee R, Mucha R, Farnell M, et al. Meningitis after basilar skull fracture: does antibiotic prophylaxis help?

Postgraduate Medicine. 1988; 83: 267-270.

42. Dagi F, Meyer F, Poletti C. The incidence of and prevention of meningitis after basilar skull fracture. American

Journal of Emergency Medicine. 1983; 3: 295-298.

43. Einhorn A, Mizrahi E. Basilar skull fractures in children: the incidence of CNS infections and the use of antibiotics.

Am J Dis Child. 1978; 132: 1121-1124.

44. MacGee E, Cauthene J, Brackett C. Meningitis following acute traumatic cerebrospinal fluid fistula. J Neurosurg.

1976; 33: 312-317.

45. Klastersky J, Sadeghi M, Brihaye J. Antimicrobial prophylaxis in patients with rhinorrhea or otorrhea: a double-

blind study. Surg Neurol. 1976; 6: 111-114.

46. Calvert C, Cairns, H. Discussion on injuries of the frontal and ethmoidal sinuses. Proceedings of the Royal Society

of Medicine. 1942; 35: 805-810.

47. Yilmazlar S, Arslan E, Kocaeli H, et al. Cerebrospinal fluid leakage complicating skull base fractures: analysis of

81 cases. Neurosurg Rev. 2006; 29: 64-71.

48. Zrebeet H, Huang P. Prophylactic antibiotics in the treatment of fractures at the base of the skull. Del Med Jrl.

1986; 58: 741-748.

49. Villalobos T, Arango C, Kubilis P. Antibiotic prophylaxis after basilar skull fractures: a meta-analysis. 1998; 27:

364-369.

50. Ignelzi R, VanderArk G. Analysis of the treatment of basilar skull fractures with and without antibiotics. J

Neurosurg. 1975; 43: 721-726.

Davis 20

Appendix A

Glasgow Coma Scale (GCS)

Score Eye opening Verbal response Motor response

1 None None No movements

2 Open to painful stimulation

Incomprehensible sounds Extends to pain

3 Open to voice Inappropriate words Abnormal flexion to pain

4 Open spontaneously Confused, disoriented Withdrawals to pain

5 - Oriented, converses Localizes to painful stimulus

6 - - Obeys commands

Severity of TBI based on GCS

Severity GCS Score

Mild 13-15

Moderate 9-12

Severe <8