xRead - September 2022

Cerebrospinal fluid rhinorrhea management

The literature search identified 19 studies that met in clusion criteria and were pertinent to the use of lumbar drainage following endoscopic repair of CSF leaks (see Ag gregate grade of evidence below for breakdown). Many of these case series report use of lumbar drains in a significant portion of their patient population with successful endoscopic repair rates ranging from 83% to 94% on first attempt 10,20–24 and 91% to 100% on sec ond attempt. 10, 20, 21, 23, 24 As a result, some have concluded that lumbar drains are necessary adjuncts to achieve these success rates. On the other hand, some studies indicate otherwise. In a meta-analysis assessing a total of 289 CSF fistulae, Hegazy et al. 4 reported a similarly high success rate of 90% on first repair attempt, but found no statistically significant difference in this outcome whether or not lum bar drain was used. Similarly, in a randomized controlled trial, Albu et al. 25 attained a 92% to 95% success rate of endoscopic repair and saw no association between this suc cess and lumbar drain use ( p = 0.2). Additional case series had similar findings despite the etiology of the leak 26 or the size of the dural defect, 27 but did identify a difference in the associated length of hospital stay. This difference was reported both in patients managed conservatively for traumatic CSF rhinorrhea with lumbar drain and in those with lumbar drain following endoscopic repair of CSF leak, when compared to their respective counterparts managed without lumbar drain. 24, 28, 29 Ransom et al. 30 noted lum bar drain complications led to readmissions and additional days in the hospital, and occurred more frequently than the recurrent CSF leaks they were proposed to prevent. In an institution-specific study, Zuckerman and DelGaudio 29 reported that the use of a lumbar drain was associated with a mean additional 3.8 hospital days. However, studies have indicated that spontaneous CSF leaks are a variant of benign or idiopathic intracranial hy pertension, and following endoscopic repair patients typi cally continue to have elevated intracranial pressures, pur portedly placing them at risk for a higher rate of treatment failures. 31–35 In this setting, lumbar drains have been used to measure postoperative intracranial pressures in order to identify patients that may benefit from CSF pressure lowering treatments, including acetazolamide and ventricu loperitoneal or lumboperitoneal shunts, as it may improve endoscopic repair outcomes. 31, 36, 37 In summary, the available evidence, although limited, suggests that lumbar drains do not contribute to successful repair. When weighing the benefits and harm, using lumbar drains should be thoughtfully considered because their use appears to significantly lengthen hospital stay and incur un necessary costs and potential complications. However, use in spontaneous CSF leaks may provide useful information regarding intracranial hypertension, which may necessitate further management (Supporting material 2).

Benefit: Procurement of CSF pressure measurements in pa tients with intracranial hypertension. Harm: Meningitis, persistent leak, low-pressure spinal headache, local infection, pain, invasive procedure, re tained catheter fragment, pneumocephalus, brain hernia tion. Cost: High secondary to increased length of stay and acuity of care. Benefits-harm assessment: Significant risk with some evi dence of benefit in patients with intracranial hypertension. Value judgments: None. Recommendation level: Recommendation against routine use, option for patients suspected of having underlying intracranial hypertension. Intervention: Routine use of lumbar drainage following CSF leak repair. Endoscopic repair There are a variety of different graft materials avail able for endoscopic repair of CSF leaks, including fat, bone, allografts, free mucosal grafts, and vascularized grafts, as well as glues or sealants to secure these ma terials in place. Following an extensive literature search on the topic of endoscopic repair for CSF rhinorrhea, the overall success rate ranges from 70% to 100% on first attempt 4–7,10,20–24,27,39–74 and 86% to 100% on sec ond attempt 4, 5, 10, 20, 21, 23, 24, 27, 45, 46, 57,63–65,67, 70, 74 for all reported repair materials. We conducted a review of the available literature to determine if any of these methods are superior with supporting literature. Fat grafts A literature search revealed 8 studies that included fat graft ing and met our inclusion criteria. One of these is a meta analysis and the remainders are case series. The success rates for endoscopic repair reported in these studies ranges from 33% to 100% on first repair attempt 4, 10, 23, 24, 53, 67 and 96% to 100% on second attempt. 24, 67 In the study by Mao et al., 23 21 patients underwent repair, 3 of which were by fat obliteration. Two of these failed, giving the significantly lower success rate of 33%, although limited by a very small sample size. The rest of the studies ranged from 82% to 100% on first attempt. In a meta-analysis, Hegazy et al. 4 noted no significant difference between fat grafting, which was used for 19% of the patients, and any other repair technique. When assessing repair of intraop erative leaks following sellar or extrasellar excisions, Kong et al. 58 reported no difference between repair methods for low-flow CSF leaks; however, for high-flow leaks, a tighter closure with fascia overlay and iliac bone buttress (gasket seal method) provided significantly better control of leaks than fat grafting. Studies suggest that for the majority of skull-base defects, fat grafting is a good option for repair (Supporting material 3). Aggregate grade of evidence: C (Level 2: 1 study; Level 4: 6 studies).

Aggregate grade of evidence: C (Level 1: 1 study; Level 2: 2 studies; Level 4: 11 studies).

International Forum of Allergy & Rhinology, Vol. 6, No. 1, January 2016

19