2018-19 Section 7-Neoplastic and Inflammatory Diseases of the Head and Neck eBook

a significant increase in length of stay with delay in surgi- cal drainage in both of our study populations. Just 13.6% of adult patients and 4.6% of pediatric patients who underwent surgical drainage on the day of admission went on to require hospital stays of 7 days or more, com- pared to 64.7% and 34.5% who underwent surgery on admission days 3 to 7, respectively. This important differ- ence affects patients, clinicians, and insurers. The possible reasons for different behavior in adults and children are multiple. It is thought that deep neck infections in adults arise primarily from direct inocula- tion from trauma or direct extension of infection from adjacent structures, whereas in children these infections more typically represent suppurative changes in a lymph node after infection in the sinuses, nose, adenoids, ton- sils, or middle ear. 24–27 The clinical relevance of this dif- ference is that adult infections are located along a fascial plane, with the immediate potential to spread vertically, whereas pediatric infections tend to be con- tained with a lymph node and do not spread until late, following rupture of the node. 15 The tendency for pediat- ric deep neck infections to spread later in the disease course could explain in part why abscess-specific compli- cations were less common in pediatric patients who had delay in drainage. Previous studies have identified fac- tors including younger age, smaller abscess size, and lack of fluctuance as predictors of success of medical therapy alone. 28,29 In pediatrics, abscess size > 2 cm 8 and 2.2 cm 30 have been found to be strongly associated with medial treatment failure and may help to identify which patients should undergo surgical drainage. How- ever additional studies are needed to further delineate the population of pediatric patients who require surgery. In addition to differences in deep neck abscess by age and timing of surgery, we also identified significant differ- ences based on the advanced ASA classification, preopera- tive sepsis, preoperative white blood cell count, congenital malformations, and gender. Although many of these differ- ences are well known to be associated with more severe infection, the gender differences are surprising. We found that deep neck abscess was more common in males because 57% of pediatric and 62% of adult’s patients were male. This has also been found for other respiratory infections; a review of 84 studies on gender differences found that males were more commonly affected. 31 Interestingly, we also found that females were at an increased risk for abscess specific M&M on multiple regression analysis in adults— and there was a trend in pediatrics, although this was not significant. This finding is in contrary to prior studies of gender differences in respiratory infections, which have shown that males are more likely to suffer complications. 31 The etiology of our observed gender difference in deep neck abscess is unclear and may reflect social, economic, or physiologic causes. Unfortunately, we are not able to deter- mine an etiology for the gender difference from our data. When interpreting our study, one should consider sev- eral limitations. First, our study shares the limitations of all multiinstitutional database studies, including errors or inconsistencies in coding. However, we feel that these limi- tations are mitigated by the rigorous methodology used by ACS–NSQIP and ACS–NSQIPP. Second, there are several

show that, for adult patients, there is indeed an associa- tion with delay in drainage and increased M&M; how- ever, this was not observed in pediatric patients. Deep neck abscesses have the ability to spread and cause upper airway obstruction, mediastinitis, pneumonia, sepsis, epidural abscess, or death. 14 In an effort to prevent these potential complications, many authors advocate urgent surgical drainage of deep neck abscess. 14–16 How- ever, the conventional treatment of deep neck abscess with antibiotic therapy and urgent drainage has been chal- lenged following numerous studies, suggesting that select pediatric patients improve with antibiotics alone. 10,17–19 In pediatrics, a systematic review (8 series, n 5 94) found that 52% of children resolved with medical therapy alone. 2 Sim- ilar trials have been attempted in adults. Plaza Mayor et al. described 31 adult cases treated medically, with sur- gery reserved only for medical failures. In this series, only three patients went on to require I&D. 17 An important limitation of these studies is that many diagnoses based on radiographic studies are in fact earlier stage infections. For example, ring-enhancement is frequently presumed to rep- resent encapsulation of an abscess, yet early-stage fibrosis of a phlegmon can appear very similarly. Unlike abscesses, phlegmons have an intact vascular supply that can allow antibiotics and host defenses to access the infection site, whereas abscesses with a dense fibrotic capsule and limited vascular supply may instead prevent antibiotic therapy from reaching the site of infection. 20 The sensitivity and specificity of imaging for distinguishing deep neck abscess versus cellulitis varies widely from 43% to 89% and from 0% to 63%, respectively, 5,15,21,22 including wide-ranging estimates based on the deep neck space involved. One con- sistent trend is that the sensitivity of imaging improves with increasing extent of infection. 21 When only the retro- pharyngeal space is involved, the positive predictive value of imaging for abscess is 50%; however, when more than one deep neck space is involved, the positive predictive value increases to 91%. 23 An additional limitation of the existing literature is the low overall power, which makes detection of clinically significant differences unlikely. Finally, one of the proposed benefits of conservative ther- apy is avoidance of iatrogenic injury. This risk may be over- stated because the present studies identified a rate of only 0.3% for postoperative nerve injury. To study if a delay in surgical drainage was associated with increased M&M, patients were grouped based on day after admission for which they underwent surgical drain- age of an abscess. Multivariate logistic regression was used to control for confounding factors including patient charac- teristics, comorbidities, preoperative sepsis, and degree of preoperative leukocytosis. After adjustment, we found that delay in surgical drainage was associated with significantly greater 30-day abscess-specific M&M in adults, whereas in children there were no differences between groups. The sensitivity analysis showed that these findings were robust and did not vary significantly when group definitions were modified. Interpretation of other specific complications including neurologic complications was not possible because event rates were low and prone to imprecision. Delay in surgical drainage may not only affect clini- cal outcomes, it may also prolong hospital stays. We show

Laryngoscope 126: August 2016

Cramer et al.: Delay in Drainage of Deep Neck Abscess

216