Portal de Periódicos da CAPES

ED Considerations in the Diagnosis and Treatment of Peritonsillar Abscess

2001; Lippincott Williams & Wilkins; Volume: 23; Issue: 3 Linguagem: Inglês

10.1097/00132981-200103000-00007

1552-3624

James R. Roberts,

Trauma Management and Diagnosis

A well known complication of bacterial pharyngitis that may present to the ED is a peritonsillar abscess, also termed quincy. A similar condition, peritonsillar cellulitis, was reviewed last month, and can be viewed as an abscess about to happen. Together, the two conditions have been referred to as peritonsillitis or peritonsillar sepsis. In the ED, it is important to make the diagnosis of a peritonsillar process vs. the more common simple tonsillitis/pharyngitis because peritonsillitis may not resolve with antibiotics alone, close follow-up is mandatory, and occasionally serious sequelae are seen. George Washington is said to have died from quincy. Patients may present primarily with peritonsillitis, or it may occasionally evolve over a few days despite concurrent treatment with seemingly effective antibiotics for the organisms causing the initial acute pharyngitis. Peritonsillar abscess is much less common than before the penicillin era, and the logical conclusion is that prompt antibiotic therapy will lessen its incidence. It's clear, however, that not all abscesses can be prevented with early antibiotic treatment. Patients with peritonsillitis are miserable and appear quite ill, and usually it's evident to everyone, including the patient, that this is not a run-of-the-mill viral sore throat. The purpose of this month's discussion is to highlight the pathophysiology of peritonsillar abscess, and to present a reasonable clinical approach for the emergency physician. Management of Peritonsillar Abscess, Maharaj D, et al, J Laryngol Otol, 1991;105:743 This prospective randomized clinical trial is one of a number of reports that have compared the safety and efficacy of needle aspiration vs. incision and drainage in adult patients with confirmed unilateral peritonsillar abscess. It also investigated the need for routine hospitalization, the need to treat with intravenous (rather than oral) antibiotics, and the optimal antibiotic regimen. The South African authors note that bacterial infection of the tonsils can spread to the peritonsillar space that lies between the fibrous capsule of the tonsil and the superior constrictor muscle of the pharynx. Unchecked infection results in a collection of pus, usually adjacent to the upper or medial pole of the tonsil. All authorities agree that the definitive treatment of an established abscess consists of some type of surgical drainage and antibiotics, but the literature has been divided on the need for routine hospitalization, specific antibiotic protocols, and the optimal surgical technique. This study of 60 patients (ages 15 to 50 years) was randomized to evaluate drainage by needle aspiration or traditional I & D. To confirm the presence of pus, all patients initially underwent permucosal needle aspiration at the point of maximum fluctuation. If pus was obtained, the remainder of the purulence was drained by continued needle aspiration alone or by formal I & D. I & D was done with a guarded scalpel blade, and aspiration was performed with a 10 ml syringe/18 gauge needle. Antibiotics consisted of an initial IM dose of procaine penicillin followed by 500 mg of oral penicillin VK every six hours for 10 days. All patients were treated as outpatients with follow-up on post-operative days one and seven. If pus reaccumulated, it was drained again with the original procedure. Of 77 patients with suspected peritonsillar abscess, pus was aspirated in 60 (78%). All presented with odynophagia and drooling, and about half had trismus, but fever was generally mild. Streptococcus (pyogenes and others) was cultured in 62 percent of cases, but anaerobes also were commonly isolated. In 15 percent of cases, no organism was isolated, and there were two cases of E. coli (sensitive to penicillin) and one case of Klebsiella. Only 82 percent of the cases returned for follow-up on the first post-operative day. The initial success rate was similar with both techniques: 87 percent with needle aspiration and 90 percent with I & D. A repeat procedure was required in 13 percent of the needle aspiration group and 10 percent of the I & D group. Only two patients subsequently had to be admitted to the hospital, both for dehydration. Follow-up at seven days was only 63 percent, but at that time, all had resolution of the abscess. The authors believe that most patients with peritonsillar abscess can be successfully and safely treated with needle aspiration, oral penicillin, and outpatient follow-up. Although this is obviously a cost-effective protocol, the needle aspiration technique also is easier to perform than an I & D, it can be done by non-surgeons, it is generally well tolerated, and the aspiration provides almost immediate relief of symptoms in most cases. With regard to technique, the authors performed needle aspirations at three locations before they deemed pus to be absent. The needle was first inserted at the point of greatest fluctuation (which was usually the upper pole), then the mid-tonsillar region, and finally the lower pole. The authors point out that if only a single aspiration is done, one may miss the collection of pus, and usually this occurs with an isolated lower pole abscess. No patients developed airway compromise or other significant complications. Comment: The most common deep infection of the head and neck is a peritonsillar abscess. In the early 1980s, it became evident that the historical treatment of peritonsillar abscess, which consisted of hospitalization and drainage via a large incision with placement of drains, was unnecessary in the vast majority of cases. Also it was previously recommended that patients with a peritonsillar abscess were always candidates for tonsillectomy, but this is now controversial and even refuted by some. From a clinical standpoint, severe exudative tonsillitis and peritonsillar abscess may be quite similar. Therefore, all ill patients should be examined with the intention of ruling out an abscess. It is important to make the differentiation because an abscess almost always requires some sort of surgical drainage. This is an infectious process of both children and adults, but young children seem to fare better with nonsurgical antibiotic therapy than do adults. Adults usually require drainage for cure. Treating adults with a true peritonsillar abscess with antibiotics alone is usually unsuccessful. At some point in time, tonsillitis spreads past the border of the tonsil and invades the soft tissue of the peritonsillar space, probably beginning as a cellulitis. When an abscess is obvious, the treatment is standardized, but a significant number of patients present in the cellulitis phase. This discussion will focus on peritonsillar abscess, but an approach to peritonsillar cellulitis was discussed in detail last month. It was mentioned that the clinician's clinical acumen is correct in the differentiating of abscess from cellulites 70 to 80 percent of the time. For young children or in complicated cases, diagnostic CT or intraoral ultrasound has been recently advocated. Clinically, severe exudative tonsillitis and peritonsillar abscess may be quite similar, and all ill patients should be examined to rule out abscess Peritonsillar abscess is primarily a disease of young adults. They usually have co-existent bilateral exudative tonsillitis, but the peritonsillar abscess itself is almost always a unilateral disease. Severe odynophagia, fever, tender adenopathy, and trismus (from pterygoid muscle inflammation) are commonly seen in addition to a severe sore throat. A change in voice (“hot potato”), drooling, and fullness in the ears related to soft palate swelling and occlusion of the Eustachian tube may be noted. In the advanced stages, the uvula is deviated by the bulging infected mass. The most devastating complications of peritonsillar infection are airway problems (usually only in children), life-threatening deep space infections in the neck, or mediastinitis. All complications are potentially lethal, and can occur with surprising rapidity. Mediastinitis, even in previously healthy patients who have received appropriate antibiotics and needle aspiration, is rare but has been described (J Laryngol Otol 1996;110:175). It's difficult for an abscess to occur in patients who have had tonsillectomy, but it is a well known complication of simple streptococcal tonsillitis in otherwise healthy individuals. Exactly why some unfortunate patients progress to abscess formation is unknown. It is generally assumed that early antibiotic therapy will decrease the incidence of abscess formation, but in one study, 20 percent of patients with peritonsillar abscess were already taking penicillin at the time of diagnosis (Arch Otolaryngol Head Neck Surg 1987;13:984).Table: Bacteriology of Peritonsillar Abscess in a Series of 62 PatientsBecause the abscess results from a local extension of a tonsillar infection, the organisms causing peritonsillar abscess parallel those causing tonsillitis. Group A beta hemolytic strep, Streptococcus viridans, and anaerobes (Bacteroides and Fusobacterium species) are consistently recovered, and are usually exquisitely sensitive to penicillin. Even though it is an abscess in the strict sense, staphylococci are not common culprits. In a study by Ophir et al (Arch Otolaryngol Head Neck Surg 1988;14:661), not a single case of Staphylococcus aureus was found in 62 patients. Others have reported various coagulase-negative staph species, but their role in infection is unclear. Initial treatment is somewhat controversial, but most authors and multiple studies have shown that needle aspiration in the ED is a safe diagnostic and therapeutic procedure for adults with peritonsillar abscess. The accompanying table outlines the current standard approach. Young children always require special attention. This process has been described in the first year of life, and initial hospitalization is prudent in such patients. One recent study of 102 pediatric patients (age 8 months to 19 years) with possible peritonsillar abscess studied a protocol that eschewed immediate surgery, needle aspiration, or computed tomography for the first 24 hours (Laryngoscope 2000; 110:1698). Treatment consisted of IV hydration and antibiotics alone. Half of the patients (52 of 102) did well within 24 hours, and were discharged. These cures were almost all in patients under age 6, and such a rapid response without surgery in the younger patients has been noted in other studies. The other half (50 of 102) underwent tonsillectomy as therapy for the abscess. This is a very reasonable clinical approach, but because pretreatment CT/ultrasound were not performed, it is speculative whether the response to therapy correlates with the type of infection. Peritonsillar aspiration is a technique well suited for the ED. Usually the only complications are minimal bleeding and some discomfort to the patient. The rare complication of potential major significance is puncture of the carotid artery or jugular vein. This complication should not occur with a few common sense precautions, and I could not find a single reported case of this occurring with the needle aspiration technique. I prefer to use a 10 ml syringe and a short one-inch 18 gauge needle. You will never remove 10 ml of pus (usually 4–5 ml is the maximum), but the larger syringe allows good suction and affords better access to the posterior pharynx than do smaller syringes. The carotid artery and jugular vein are lateral to the tonsils and quite posterior to them. Therefore, the needle should be directed either straight back parallel to the sagittal plane or slightly medial. If pus is present, it will be encountered within the first 2 cm of the surface, usually more superficial, so penetration of greater than 2 cm is discouraged. Using a short one-inch 18 gauge needle offers a safety margin, but some operators prefer to impale a rubber stopper from a red top tube on the base of a 1 1/2 inch needle to limit the chance of advancing the needle too far. Needles smaller than 18 gauge are counterproductive. Pus usually first collects at the superior aspect of the peritonsillar space, but when advanced, the abscess can cover the anterior, lateral, and superior surfaces of the tonsils, occasionally completely obliterating the tonsil. Most authors suggest routinely performing three separate aspirations — superior, middle, and lower — before calling it quits. If you cannot identify an area of obvious fluctuation, always aspirate the superior pole first. Prior to aspirating, I find it helpful to palpate the area with my finger to identify the area most likely to yield pus. Importantly, do not aspirate the tonsil itself; the pus collects only in the peritonsillar space. If pus is obtained, as much as possible should be aspirated on the initial attempt. Once pus is encountered, other areas need not be aspirated. With attention to some minor details, the procedure is relatively easy and not as painful as one might think. In a very anxious or severely symptomatic patient, the judicious use of intravenous narcotics and/or sedation prior to drainage should be routine. I am liberal about this so everybody gets something parenterally. I prefer to begin with 2–3 mcg/kg of fentanyl intravenously about five minutes prior to the procedure. Fentanyl is a very effective analgesic that does not cause vomiting or hypotension, and it has a very short half-life (about 30 minutes). I don't advocate true conscious sedation for this procedure, but a small dose of midazolam may be cautiously added to a narcotic. Some recommend nitrous oxide. Peritonsillar aspiration is a technique well suited for the ED, with the only complications minimal bleeding and some discomfort to the patient One begins by spraying the mucosa with a topical anesthetic; 10% lidocaine spray is a good choice. In addition, I infiltrate the area to be aspirated with 1% lidocaine with epinephrine using a 25 gauge needle. Anesthetize only the mucosal surface, and be careful to avoid filling the abscess cavity with lidocaine; the pressure will cause more pain. Mucosal infiltration may be considered excessive by some, but I have found that it makes the actual needle aspiration almost totally painless and epinephrine limits bleeding. If the anesthetizing needle tip is in the correct submucosal plane, only one stick is required to anesthetize a liberal area, enough to cover additional aspirating sites. Simply continue to inject until most of the peritonsillar area blanches. Obviously a good light is required, preferably an ENT head lamp, with two tongue blades side by side to keep the tongue down. It helps to have an assistant pull the cheek laterally. The patient should be seated in an upright position — a dental chair works great — and the head should always be flat against the back of the stretcher or head rest. I keep a tonsillar suction device at the bedside. There is always some bleeding after the procedure, but it is self-limited in the patient with normal coagulation, an important concept to address before aspiration. Obviously, you wouldn't try this procedure on someone who is on Coumadin or has a coagulopathy. If pus is obtained, it probably should be sent for culture, although this is not a universal standard. There are generally few laboratory surprises, and treatment is rarely changed by the results. Failure to grow organisms usually indicates fastidious anaerobes. There have been no suggestions that Chlamydia, Mycoplasma, or viruses cause peritonsillar abscess. A lateral neck x-ray, blood cultures, and a CBC offer little additional information, and they can be withheld unless there are extenuating circumstances.Table: Some Physical Findings in Peritonsillar Abscess: A Series of 62 PatientsIt is universally agreed that antibiotics should be administered. The literature strongly supports the initial use of penicillin, and surprisingly excellent cure rates are reported with oral penicillin in modest doses of about 500 mg QID. It's difficult for me to give only oral medications for patients with this disease, so I always give an initial dose of a parenteral antibiotic, usually intravenous penicillin G. It's may be overkill, but I prescribe 5 million units for this loading dose along with a gram of benemid (Probenecid) to decrease renal excretion of the antibiotic. Erythromycin is usually recommended for patients with penicillin allergy, but I think that's a very bad choice in this setting. The last thing patients need on top of their severe sore throat is the GI upset that often accompanies any form of erythromycin. In my mind, clindamycin is the ideal alternative to penicillin (but I would not argue with using the newer macrolides, Biaxin or Zithromax). A first-generation cephalosporin in the absence of previous anaphylaxis to penicillin is another good choice. Because anaerobes can be cultured, some authors add metronidazole to this regimen. Ampicillin/sulbactam (Unasyn) was no better than procaine penicillin in one study (Eur Arch Otorhinolaryngol 1998;225:163). Decadron (10 mg IM/IV) is an unstudied but commonly administered adjunct. Once the aspiration has been performed, I usually watch the patient for about two hours and discharge only those who are clinically non-toxic and presumed reliable. Those who look ill, are immunocompromised, or who I think probably won't take their pills or comply with follow-up should be admitted. For outpatients, I insist on a 24-hour follow-up for all. Intravenous hydration, fever control, and a narcotic during the ED observation period make patients feel a lot better, and removing only a few cc of pus remarkably reduces pain. A few narcotic tablets to go also are mandated. Gargling with warm saline seems to do a good job to reduce pain, and outpatients should do this every two hours. A small amount of bleeding may persist for 12 to 24 hours, and patients often relate the sensation of swallowing pus for the first day (an unpleasant thought but presumably a desirable clinical sign of continued drainage). The transmucosal needle aspiration technique and outpatient management of an established peritonsillar abscess is well supported in the literature. Ophir et al (Arch Otolaryngol Head Neck Surg 1988;114:661) evaluated needle aspiration and oral antibiotics in 115 patients. If pus was obtained, subjects were treated with oral amoxicilin 500 mg TID. Those with peritonsillar cellulitis were admitted to the hospital (not a universal standard), and those who exhibited systemic toxicity also were admitted (a given). In 85 percent of patients, the abscess resolved completely with aspiration as the sole surgical intervention. The authors suggested a somewhat different approach by admitting all patients with a negative aspiration to the hospital. I could find no general consensus to support routine admission for patients with a negative aspiration, and some use this as an indication of cellulitis and a criterion for discharge. Ophir et al also question the need for routine culture and sensitivity because in no patient did the results affect the outcome or change therapy. I agree with that observation. Spires et al (Arch Otolaryngol Head Neck Surg 1987;113:984) likewise found that needle aspiration, outpatient follow-up, and oral antibiotics usually produced a successful outcome. The recent medical literature clearly supports needle aspiration as an acceptable initial treatment for peritonsillar abscess. Outpatient therapy with oral antibiotics in the nontoxic, nonimmunocompromised, reliable patient is well accepted. The first follow-up should occur within 24 to 36 hours. At some point in follow-up, all patients should probably see an otolaryngologist for evaluation for tonsillectomy. Although tonsillectomy used to be considered mandatory for peritonsillar abscess, this delayed intervention is currently controversial. However, peritonsillar abscess will recur in up to 20 percent of cases so the old adage “No tonsil equals no peritonsillar abscess” may cause some to consider surgery. Before the emergency physician embarks on the strategy outlined above, cooperation should certainly be elicited from the ENT department. A few treatment failures of needle aspiration will surely occur, toxicity can worsen, and reliability of the patient can be misjudged so there must be a smooth transition from outpatient to inpatient therapy. Approximately 10 percent of the time, any form of surgical drainage will need to be repeated. Peritonsillar abscess is primarily a disease of young adults, and the abscess itself is almost always a unilateral disease Needle aspiration, a common intervention in adults, also has been studied in children as outpatients. Adults are difficult enough to talk into needle aspiration (you're going to put that needle where?), but children are impossible. ED conscious sedation with midazolam, ketamine, and glycopyrrolate has been advocated as a safe and efficacious approach to I & D of peritonsillar abscess (Arch Otolaryngol Head Neck Surg 1999;125:1197). Such an approach in a young child requires a team effort in a milieu with appropriate time, space, and nursing help. I would not attempt this in the middle of the night by myself. IV narcotics are mandatory for this procedure, but full conscious sedation requires more resources. Finally, a small percentage of patients will have coexistent mononucleosis and a peritonsillar abscess. The relationship is unclear (In Med J 1999;92:278). Consensus Guideline for the Treatment of Suspected Peritonsillar Abscess in Adults ▪ Needle aspiration as initial surgical drainage ▪ Treatment on an outpatient basis ▪ Penicillin is the antibiotic of choice if not allergic ▪ Adequate pain medication ▪ No benefit of routine culturing of pus Note: These guidelines are based on a cohort study of 123 patients, a national survey of otolaryngology practice, and a meta-analysis of various components in the treatment regimen. Source: Laryngoscope 1995;105:1.