| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
State of the Art |
1 From the Department of Radiology, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104 (B.A.B.); and the Department of Medical Imaging, the Toronto Hospital, General Division, Ontario, Canada (S.R.W.). Received March 10, 1999; revision requested May 3; revision received June 29; accepted July 22. Address correspondence to B.A.B. (e-mail: birnbaum@oasis.rad.upenn.edu).
Abstract
Acute appendicitis is a common clinical problem. Accurate and prompt diagnosis is essential to minimize morbidity. While the clinical diagnosis may be straightforward in patients who present with classic signs and symptoms, atypical presentations may result in diagnostic confusion and delay in treatment. Helical computed tomography (CT) and graded compression color Doppler ultrasonography (US) are highly accurate means of establishing the diagnosis. These imaging modalities have now assumed critical roles in the treatment of patients suspected to have appendicitis. The purpose of this article is threefold: to provide an update on new information regarding the pathophysiology, clinical diagnosis, and laparoscopic treatment of acute appendicitis; to describe the state-of-the art use of CT and US in diagnosing this disease entity; and to address the role of medical imaging in this patient population.
Index terms: Appendicitis, 751.291 • Appendix, CT, 751.12112, 751.12115 • Appendix, US, 751.12983 • State of the Art
Appendicitis is the most common cause of acute abdominal pain that requires surgical intervention in the Western world (1). Patients with the disease may present with a wide variety of clinical manifestations, and the diagnosis may elude even the most experienced clinicians (2). Prompt diagnosis is essential to minimize morbidity, which remains substantial if perforation occurs. The advent of antibiotics and effective surgical management have substantially reduced appendicitis-related mortality; however, deaths from appendicitis still occur, particularly in the elderly.
Appendicitis was rare in the past and remains so in underdeveloped countries (3). There appears to be no record of early physicians, from Hippocrates to Moses Maimonides, recognizing this disease entity (3). Although the anatomy of the appendix was well known by the 18th century, it was not until this time that it was recognized that the appendix could become inflamed, with possibly fatal consequences (4). Early reports of perityphlitis and typhlitis in the 19th century appeared to describe a new clinical phenomenon (3,4). Confusion over this right-lower-quadrant entity existed until Reginald H. Fitz presented his landmark article in 1886, in which he coined the term "appendicitis" and correctly classified this disease by describing the appendix as the primary source of inflammation in acute typhlitis (5). Fitz described the signs and symptoms of acute and perforated appendicitis, outlined the progression from acute right-lower-quadrant inflammation through peritonitis and iliac fossa abscess formation, and recommended early appendectomy if there were signs of spreading peritonitis or of clinical deterioration. Shortly thereafter, Charles McBurney and other pioneering surgeons began to intervene early in acute appendicitis (6,7). These clinicians advocated prompt clinical diagnosis and surgical intervention. Their surgical aim was to operate in a timely fashion before appendiceal perforation and peritonitis developed.
The goal of modern surgical management essentially is the same and focuses on a balance between the rate of false-negative laparotomy and the rate of perforation at the time of surgical exploration (8–10). It is tradition that surgeons have diagnosed appendicitis on the basis of patient history and physical examination results. The relatively recent introduction of new imaging technology—in particular, graded compression color Doppler US and helical computed tomography (CT)—potentially has changed "the rules of the game." The purpose of this article is to document recent advances in our understanding of appendicitis and to define the role of medical imaging in patients with this condition.
ANATOMY AND PATHOPHYSIOLOGY
The adult appendix is a long diverticulum averaging 10 cm in length that arises from the posteromedial wall of the cecum, approximately 3 cm below the ileocecal valve (11). Although the relationship of the base of the appendix to the cecum essentially is constant, the remainder of the appendix is free, which accounts for its variable location in the abdominal cavity. The appendix may lie in a retrocecal, subcecal, retroileal, preileal, or pelvic site. This variability in location may greatly influence the clinical presentation in patients with appendicitis (12,13).
Acute appendicitis may occur at any age, although it is relatively rare at the extremes of age. The maximum incidence of the disease occurs in the 2nd decade; thereafter, disease incidence declines with age (14,15). The primary pathogenic event in the majority of patients with acute appendicitis is luminal obstruction (16–18). This may result from a variety of causes, which include fecaliths, lymphoid hyperplasia, foreign bodies, parasites, and both primary (carcinoid, adenocarcinoma, Kaposi sarcoma, and lymphoma) and metastatic (colon and breast) tumors (19–26).
Fecaliths, which result from the inspissation of fecal material and inorganic salts within the appendiceal lumen, are the most common cause of obstruction and are present in 11%–52% of patients with acute appendicitis (20–22). True appendiceal calculi (hard, noncrushable, calcified stones) are less common than appendiceal fecaliths (hard but crushable concretions) but have been shown to be associated more commonly with perforating appendicitis and with periappendiceal abscess (20).
Once appendiceal obstruction occurs, the continued secretion of mucus results in elevated intraluminal pressure and luminal distention. This stimulates the visceral afferent nerve fibers that enter the spinal cord at thoracic levels T8 through T10, which causes referred epigastric and periumbilical pain (26). This visceral pain usually is mild, is poorly localized, and is 4–6 hours in duration. Anorexia, nausea, and vomiting usually follow as the pathophysiology worsens (13). Increasing intraluminal pressures eventually exceed capillary perfusion pressure, which leads to venous engorgement, arterial compromise, and tissue ischemia. As the epithelial mucosal barrier becomes compromised, luminal bacteria multiply and invade the appendiceal wall, which causes transmural inflammation. Continued tissue ischemia results in appendiceal infarction and perforation. Inflammation then may extend to the parietal peritoneum and adjacent structures, which include the terminal ileum, cecum, and pelvic organs.
Patients typically experience the classic migration of pain to the right lower quadrant at this stage (13,26). This somatic pain is continuous and is more severe than the early visceral pain. The classic migration of pain need not occur, and the point of maximal tenderness may be distinct from McBurney's point if the appendix is in an atypical location.
Patients with acute appendicitis usually are afebrile or have a low-grade fever. Perforation should be suspected whenever a patient's temperature exceeds 38.3°C (27). If perforation does occur, periappendiceal phlegmon or abscess will result if the terminal ileum, cecum, and omentum are able to "wall off" the inflammation. Peritonitis usually develops if there is free perforation into the abdominal cavity.
Mild acute appendicitis may resolve spontaneously, with or without antibiotic therapy, if the inciting obstruction is relieved (16,28–33). It is presumed that this occurs after a soft fecalith is expelled from the appendiceal lumen or if lymphoid hyperplasia is the cause of the symptoms (30).
Recurrent and chronic forms of appendicitis also have been recognized and occur with an approximate incidence of 10% and 1%, respectively (31–35). Recurrent appendicitis is characterized by a history of similar episodic attacks of right-lower-quadrant pain that lead to appendectomy, with a histopathologic diagnosis of acute inflammation of the appendix.
Diagnostic criteria for chronic appendicitis include a history of right-lower-quadrant pain of at least 3 weeks duration, no alternative diagnosis, histopathologic evidence of chronic active inflammation of the appendiceal wall or of fibrosis of the appendix, and complete relief of symptoms after appendectomy.
Nearly all patients with recurrent or chronic appendicitis are able to recall at least one episode of acute abdominal pain consistent with acute appendicitis that was managed nonsurgically. This suggests that recurrent and chronic appendicitis can be avoided with the accurate diagnosis and surgical management of acute appendicitis (32).
CLINICAL DIAGNOSIS
The clinical diagnosis of acute appendicitis is based primarily on patient history and on physical examination results. In a classic presentation, a patient with appendicitis has a typical historical sequence of symptoms (poorly localized periumbilical pain followed by nausea and vomiting, with subsequent migration of pain to the right lower quadrant) and physical findings that vary with time and with the location of the appendix (13,27). This classic presentation occurs in only 50%–60% of patients, and the diagnosis may be missed or delayed when atypical patterns of disease are encountered. Unusual presentations most likely occur when the appendix is in an atypical location (hidden from the anterior parietal peritoneum), when the patient is at an extreme of age, or when the patient is pregnant.
The most valuable elements of the patient history and physical examination were recently determined in a meta-analysis (13) of studies whose authors reported the clinical usefulness of various signs and symptoms in adults with appendicitis. A diagnosis of appendicitis was most likely in the presence of right-lower-quadrant pain, rigidity, and migration of the initial periumbilical pain to the right lower quadrant. The absence of right-lower-quadrant pain and of the classic migration of pain and the presence of similar pain in the past were shown to be historical symptoms that decreased the likelihood of appendicitis. The authors of this study (13) stressed that no single finding is able to help effectively rule out appendicitis, and they advised close follow-up of patients with abdominal pain who do not undergo further diagnostic testing.
The overall accuracy for diagnosing acute appendicitis is approximately 80%, which corresponds to a mean false-negative appendectomy rate of 20% (8,14,15,20,27,35–42). Diagnostic accuracy varies by sex, with a range of 78%–92% in male and 58%–85% in female patients. These differences reflect the fact that appendicitis may be extremely difficult to diagnose in women of childbearing age, because symptoms of acute gynecologic conditions such as pelvic inflammatory disease may manifest similarly (43,44). This diagnostic problem has led to false-negative appendectomy rates as high as 47% in female patients aged 10–39 years (41).
Diagnostic accuracy was noted to improve in the United States between 1970 and 1984, from 86% to 92% in male patients and from 74% to 83% in female patients (14). In a recent review of medical records of 4,950 patients who underwent emergency appendectomy at U.S. Department of Defense Hospitals worldwide, the false-negative appendectomy rate was noted to be 9% in male patients and 19% in female patients (36). These results suggest that diagnostic accuracy may have reached a plateau since the mid-1980s.
APPENDICEAL PERFORATION
Early surgical intervention in patients with acute appendicitis is imperative to avoid appendiceal perforation, which is associated with increased morbidity and mortality compared with nonperforating appendicitis (8,9). The overall incidence of perforation is 16%–39%, with a median of 20% (8,12–14,26,27,35–42). Perforation rates are strongly age related and are highest in the very young (40%–57%) and in the elderly (55%–70%), in whom misdiagnosis and delayed diagnosis are common.
The relationship between diagnostic accuracy and perforation remains controversial. While some authors (8,14) have reported a linear correlation between diagnostic accuracy and the rate of perforation, authors of more recent studies (37,39) have refuted any such relationship. Conventional surgical wisdom is predicated on the belief that an inverse relationship exists between the normal appendectomy rate and the perforation rate (8,9). This historical doctrine asserts that a false-negative appendectomy rate of 15%–23% is an appropriate index of management and that the failure to maintain such a surgical threshold is an indication of insufficient surgical aggressiveness and of an excessive rate of perforation. This "scare philosophy" has been questioned for some time, as it has been shown that intensive in-hospital observation can lead to improved diagnostic accuracy and to fewer false-negative appendectomies, without affecting the perforation rate (45). Moreover, recent reports (39,46) have shown no correlation between the rates of perforation and false-negative appendectomy.
The rates of perforation and false-negative appendectomy have been relatively stable over the past half century (36). This is likely due to the fact that these are two independent phenomena with their own influencing factors (46). Perforation rates correlate with time from onset of symptoms to treatment and are highly dependent on both patient-related (prehospital delay) and physician-related (in-hospital delay) variables (45–50).
False-negative appendectomy rates reflect the diagnostic difficulty encountered in differentiating appendicitis from other acute abdominal conditions. New imaging technology has the potential to alter the clinical approach to appendicitis by improving these clinical outcomes. Diagnostic evaluation with helical CT and with graded-compression US may lower the false-negative appendectomy rate because of these modalities' proved use in providing an accurate diagnosis in the overwhelming majority of patients who present with acute right-lower-quadrant pain (51). In theory, imaging in patients with equivocal clinical findings and with uncertain diagnoses also may reduce the rate of perforation by shortening in-hospital delay in treatment. These imaging examinations should be performed expeditiously to avoid incurring an increase in the perforation rate.
CT IN ACUTE APPENDICITIS
CT is a highly accurate and effective cross-sectional imaging technique for diagnosing and staging acute appendicitis (51–62). CT is readily available, is operator-independent, is relatively easy to perform, and has results that are easy to interpret. Diagnostic sensitivity and specificity are excellent for the entire spectrum of disease manifestations and do not decrease in the presence of perforation or aberrant appendiceal location. Moreover, extremes of body habitus rarely limit study acquisition or interpretation when optimized scanning methods are used.
Helical CT has reported sensitivities of 90%–100%, specificities of 91%–99%, accuracies of 94%–98%, positive predictive values of 92%–98%, and negative predictive values of 95%–100% for the diagnosis of acute appendicitis (57–62). These results are comparable to those achieved by experienced investigators (54,55) who have used thin-section, conventional, contrast material–enhanced CT and are superior to recently reported clinical accuracy (14,36).
CT Technique
All current helical CT protocols for imaging in patients suspected to have appendicitis incorporate the prospective acquisition of thin-section (
5-mm section collimation) images in the right lower quadrant. This scanning philosophy is based on the fact that CT diagnostic sensitivity and specificity are maximized only when a deliberate effort is made to image the appendix. The value of improved z-axis resolution in this clinical setting has been demonstrated with both conventional and helical CT (53,62). In a recent study, Weltman et al (62) showed that the use of 5-mm-section helical CT enabled the improved visualization of abnormal appendices (94% vs 69%), calcified appendicoliths (38% vs 19%), and periappendiceal inflammation (98% vs 75%) compared with 10-mm-thick–section helical CT in the same patient.
Appendiceal CT protocols differ considerably with regard to the anatomic area to be included in the scan and to the use of intravenously, orally, and rectally administered contrast material. The most popular and conservative approach is to perform helical CT scanning of the entire abdomen and pelvis with intravenous and oral contrast material. Proponents of this technique believe that contrast-enhanced CT is essential in the diagnosis and staging of numerous inflammatory, ischemic, and neoplastic processes that may cause acute abdominal pain and may simulate appendicitis (63).
Intravenous contrast material has been shown to aid in the diagnosis of appendicitis by permitting the identification of the inflamed appendix. This may be critical in patients with mild appendicitis and a paucity of mesenteric fat and in those with perforated appendicitis (51,52). Opacification of the terminal ileum and cecum with oral contrast material has been advocated to avoid false-positive results, in which fluid-filled terminal ileal loops are misdiagnosed as distended, inflamed appendices (53). Moreover, opacification of the normal appendix serves to exclude appendicitis. All of these factors are operative in children, which has led to the recommendation that pediatric appendiceal CT studies be performed with both intravenous and oral contrast agents (64).
Adequate enteral opacification of ileocecal bowel may take 45–60 minutes. To expedite scan acquisition, Rao et al (61) have promoted a focused appendiceal CT technique in which a limited helical CT study of the right lower quadrant is performed after the rapid administration of colonic contrast material. This technique has proved to be as accurate as those techniques in which intravenous and oral contrast material are administered, while allowing scanning completion within 15 minutes in the majority of patients examined. A limitation of this scanning method is that a minority of patients will require additional scanning of the proximal abdomen or of the distal pelvis to identify disease not included in the scanning field of view. If the initial examination results are negative, the cause of the patient's symptoms may go undetected unless the remainder of the abdomen and pelvis is imaged.
The fastest CT protocol has been promoted by Lane et al (57,58), who have advocated use of nonenhanced helical CT of the entire abdomen and pelvis. This examination may be performed in 10 minutes, does not expose the patient to the potential risks associated with iodinated contrast agents, requires no bowel preparation, and represents the most cost-effective imaging alternative to US. This procedure is most effective in patients with large body habitus, as diagnostic accuracy may be compromised in patients with little abdominal and intrapelvic fat (56,57). These investigators (58,65) have shown that nonenhanced CT is an accurate technique for establishing an alternative diagnosis in patients suspected to have appendicitis.
CT Criteria for the Diagnosis of Acute Appendicitis
Visualization of the appendix is strongly dependent on the type and quality of the CT examination, although appendiceal size, the amount of periappendiceal fat, and the degree of ileocecal bowel opacification are important influencing factors (51,52). In complicated cases, dynamic cine review of images on the CT console may facilitate recognition of the appendix, terminal ileum, and cecum. The normal appendix is identified in 67%–100% of symptomatic adults who undergo thin-section helical CT of the right lower quadrant (57,59–62,66).
Appendiceal visualization is technique dependent, with the highest detection rates reported in patients who have received rectal contrast material. When seen, the normal appendix appears as a tubular or ringlike pericecal structure that is either totally collapsed or partially filled with fluid, contrast material, or air. In our experience, the normal appendiceal wall measures less than 1–2 mm in thickness. The periappendiceal fat should appear homogeneous, although a thin mesoappendix may be present.
A definitive CT diagnosis of acute appendicitis can be made if an abnormal appendix is identified or if a calcified appendicolith is seen in association with pericecal inflammation (51–54). The appearance of the abnormal appendix varies with the stage and severity of the disease process. The CT findings are most subtle in patients with mild, nonperforating appendicitis who undergo scanning shortly after the onset of symptoms. In these patients, the appendix may appear as a minimally distended, fluid-filled, tubular structure 5–6 mm in diameter surrounded by the homogeneous fat attenuation of the normal mesentery. This appearance is seen in only the most incipient forms of acute appendicitis and, in our experience, occurs in fewer than 5% of patients who undergo scanning.
Most patients who undergo CT demonstrate greater degrees of luminal distention and evidence of transmural inflammation (Fig 1). The inflamed appendix usually measures 7–15 mm in diameter. Circumferential and symmetric wall thickening is nearly always present and is best demonstrated on images obtained with intravenous contrast material enhancement (51,52) (Fig 2). The thickened wall usually is homogeneously enhanced, although mural stratification in the form of a target sign may be noted.
|
|
|
|
|
|
|
|
The CT findings of recurrent and chronic appendicitis are identical to those of acute appendicitis (70). Distal appendicitis is diagnosed when CT reveals appendicitis that involves the distal, "upstream" aspect of the appendix, with a normal appearance of the proximal appendix and cecal apex (71). An obstructing appendicolith often is identified at the transition point between the normal and abnormal appendiceal segments (Fig 6). The proximal appendix in these cases may be collapsed or partially filled with contrast material or with air.
|
Many surgeons believe that there is little value in attempting to drain a nonliquefied phlegmon and prefer initial nonsurgical treatment with antibiotic therapy in such cases. Patients with well-defined and well-localized periappendiceal abscesses typically benefit from CT-directed percutaneous catheter drainage. In either case, interval appendectomy may be performed after an appropriate period. Patients with extensive and poorly defined collections usually require immediate surgical exploration and abscess drainage (73).
US IN ACUTE APPENDICITIS
US is a widely available and inexpensive modality with the potential for highly accurate imaging in the patient suspected to have acute appendicitis. Although operator skill is an important factor in all US examinations, it has particular importance in the examination of the patient with right-lower-quadrant pain. The learning curve required to develop the technique for scanning the right lower quadrant is considerable, and there are many pitfalls to be aware of (76). Nonetheless, the criteria for the US-based diagnosis of acute appendicitis are well established and reliable (28,51,77). In the patient in whom acute appendicitis does not explain his or her pain, US is also highly useful in identifying an alternate diagnosis (78).
In experienced hands, US has reported sensitivities of 75%–90%, specificities of 86%–100%, accuracies of 87%–96%, positive predictive values of 91%–94%, and negative predictive values of 89%–97% for the diagnosis of acute appendicitis (28,79–82). The inexperienced sonologist, working with poor equipment and/or technique, will not provide the excellent results possible with this modality.
US Technique
US examination of the patient suspected to have appendicitis should include a thorough evaluation of both the abdomen and the pelvic organs. In women in whom the answer is not evident after the performance of these two examinations, endovaginal US should be added. This is of particular importance if one considers the overlap in the symptoms of appendicitis with those of gynecologic disease in women in the childbearing years. A gynecologic explanation for the symptoms may be evident on the endovaginal images. Conversely, the appendix may have a pelvic location, in which case it may be seen clearly on the endovaginal image when it is not evident on the suprapubic image.
The specific US approach to the right lower quadrant should include graded compression US, a technique first popularized by Puylaert (77) by using high-frequency linear probes. It describes the use of uniform pressure on the region of interest by the handheld US transducer. Normal and gas-filled loops of gut will be either displaced from the field of view or compressed between the layers of musculature of the anterior and the posterior abdominal walls. In contrast, abnormal loops of gut, or the obstructed appendix, will be noncompressible and optimally seen on the graded compression image.
In addition to its benefit of showing the obstructed appendix as a noncompressible loop of gut, the technique also allows for successful examination of the patient who may have peritoneal irritation and sensitivity. If a normal, rapid US technique were used or if the transducer were repeatedly placed on the skin surface and removed, rebound tenderness would be elicited, and the uncomfortable patient would react quickly with termination of the examination.
In contrast, graded compression US, with slow and gentle maintained pressure, allows for a lengthy and successful evaluation of the area of interest in even the most uncomfortable and reluctant of patients. The patient is also able to provide input as to the point of maximal tenderness, which often is useful in focusing the examination in the correct area (83).
High-frequency linear probes for graded compression US are still in wide use today in patients suspected to have appendicitis. We have found, however, that in many patients, the currently popular curvilinear probes work equally well and provide a slightly larger field of view and greater penetration. This is of particular benefit in the obese patient in whom it is difficult to scan adequately with a poorly penetrating, linear-array, small-parts transducer. We believe that the critical factor is a transducer with a variable or short focal zone and with a frequency of 5–9 MHz.
The addition of color Doppler US also is of benefit in the evaluation of inflammatory conditions of the intestinal tract. The activity of inflammation is proportional to the amount of color signal detected within the gut wall (84,85). The normal gut is thin walled and compliant and frequently shows peristaltic activity. Hence, the detection of color Doppler ultrasound signals from the normal gut is extremely difficult.
In contrast, the thick-walled and noncompressible appendix, maintained in a fixed position by the compressing transducer, will show circumferential color when inflamed. The contribution of Doppler US is most evident in the case of the equivocal gray-scale US examination, in which it is uncertain as to whether the imaged appendix is normal or inflamed (84). Although each generation of US equipment shows improved sensitivity in the detection of Doppler ultrasound signals, inflamed gut uniformly will show greater flow than normal gut. We have found that the degree of bowel perfusion may be overestimated with power Doppler US because the power Doppler ultrasound signal is sensitive to motion artifact. We therefore prefer to use color Doppler US to detect genuine blood flow signal.
An excellent routine for the actual US examination of the right lower quadrant is to scan in the transverse plane by starting from the tip of the liver and proceeding to the pelvic brim. Several sweeps from the lateral aspect to the medial aspect may be necessary. The ascending colon usually is appreciated by its gas content and haustral pattern. In the location of the cecum, careful attention should be paid to inflammatory changes in the perienteric fat and the appendix itself. Sagittal and oblique images should then be obtained until the entire region of interest has been scanned. Detailed images are obtained of the appendix, if it is seen. We start the examination with a curvilinear transducer appropriate for the patient: a 3.5-MHz transducer for large patients and a 5-MHz transducer for thin patients. The linear transducer is used last, for more detailed images of the gut.
US Criteria for the Diagnosis of Acute Appendicitis
Rigorous adherence to the criteria for diagnosing appendicitis is recommended. The inflamed appendix is seen as a blind-ended, tubular structure with a laminated wall that arises from the base of the cecum (Fig 7a, 7b). It should be aperistaltic and noncompressible. A threshold diameter of 6 mm is invaluable for diagnosing acute appendicitis (28). Circumferential color in the wall of the inflamed appendix on color Doppler US images is strongly supportive evidence of active inflammation (Fig 7c, 7d).
|
|
|
|
|
|
With perforation of the appendix, the distended appendix may no longer be visualized at US examination. Although the criteria for the diagnosis of appendicitis are focused on the appendix itself, inflammatory changes in the perienteric fat are often the first and most obvious findings at US examination. Inflamed fat appears at US as an "echogenic mass effect." It separates the inflamed gut from the surrounding gut and other organs. The hyperemia seen in the inflamed gut extends to the inflamed fat, as seen at color Doppler US (88). Phlegmonous change manifests as hypoechoic zones with poor margination within the inflamed fat that blend imperceptibly at its margins with the fatty tissue (89). Liquefaction and abscess formation will manifest as an actual fluid component (Fig 10). Gas bubbles within a collection suggest either perforation or gas-forming organisms. A localized perforation of the appendiceal tip may show gas pockets localized to the perforation site, with disruption of the wall at that point (Fig 11).
|
|
Pitfalls in Diagnosis
The most common sources of error in the overdiagnosis of appendicitis with US include misinterpretation of the terminal ileum as the appendix and misinterpretation of a normal appendix as an inflamed appendix. The terminal ileum, in contrast to the appendix, does not attach to the base of the cecum, is not blind-ended, and shows frequent peristaltic activity. Also, the terminal ileum usually is oval in cross-section as compared with the appendix, which is round.
The normal appendix is seen infrequently at US, although it may be seen, particularly in thin patients, with excellent-quality examinations. Rioux (90) described the visualization of the normal appendix in an amazing 102 of 125 (82%) patients without acute appendicitis. In our experience and in that of others (28,82), this number usually is substantially lower, 0%–4%, in the adult population, regardless of technique. The threshold diameter of 6 mm, above which inflammation is present, is invaluable in distinguishing the normal appendix from the inflamed appendix, as the diameter of the normal compressed appendix is invariably less than this (28). Color Doppler US also is essential in distinguishing the normal from the inflamed appendix, as the hyperemia of inflammation will not manifest in the normal structure (84).
The spontaneous resolution of appendicitis is a relatively uncommon but documented condition that is a component of overdiagnosis (30,91). Affected patients may show true features of appendicitis at US, but their clinical condition favors conservative treatment, during which their symptoms subside. These patients are not encountered with any frequency, and follow-up examination to show the return of US features to normal is appropriate.
The underdiagnosis of appendicitis is much more difficult to address. Technique is of the utmost importance, as is adherence to the diagnostic criteria. Poor-quality examinations undoubtedly account for some diagnostic errors. If inflammation is localized to only the tip of the appendix, an incomplete examination also could lead to an erroneous impression that the appendix is normal. Lim et al (92) emphasized the requirement of visualizing the entire length of the appendix to avoid a false-negative diagnosis. Other problems may be related to a position of the appendix that makes it more difficult to appreciate, in particular when it is in the true pelvis and when it is retrocecal. Furthermore, perforation of the appendix may lead to decompression of the appendiceal lumen, such that the appendix, per se, is no longer seen. This removes the specificity of the US study and also may lead to diagnostic errors.
Interpretation of US findings in the patient with inflammatory bowel disease, especially Crohn disease, also may be difficult at times. The appendix may be involved in the inflammatory process of Crohn disease, or, conversely, appendicitis may be the first manifestation of this disease (93,94).
US VERSUS CT
Patients with clinical signs and symptoms typical of acute appendicitis undergo immediate surgery without radiologic evaluation. Radiologic imaging usually is requested in patients with atypical or confusing clinical findings. The choice between US and CT in this clinical setting largely is dependent on institutional preference and on available expertise, although patient age, sex, and body habitus are important influencing factors.
US is rapid, noninvasive, inexpensive, and requires no patient preparation or contrast material administration. Because US involves no ionizing radiation and excels in the depiction of acute gynecologic conditions, it is recommended as the initial imaging study in children, in young women, and during pregnancy. CT represents an excellent diagnostic alternative for all other patients. CT is complementary to US and is recommended whenever US results are suboptimal, indeterminate, or normal in patients with acute abdominal pain. US is also complementary to CT and may be particularly useful in thin patients in whom the results of initial CT, no matter how it is performed, are equivocal.
An important limitation of US is that the sensitivity and specificity for perforated appendicitis are lower than those typically seen for nonperforated appendicitis. A noncompressible appendix may be identified in only 38%–55% of patients with perforation (95,96). The US-aided diagnosis of perforated appendicitis depends on the identification of secondary findings, which, in combination, provide a specificity of only 60% (87). CT is preferred in patients suspected to have appendiceal perforation because diagnostic accuracy remains high and because CT is particularly useful for characterizing periappendiceal inflammatory masses.
In the only prospective study to our knowledge to date in which these modalities are compared, Balthazar et al (54) showed CT to be superior to graded compression US in the diagnosis of acute appendicitis. Analysis of the data for CT and US revealed similar specificities (89% vs 91%, respectively) and positive predictive values (96% vs 95%, respectively); however, CT demonstrated higher sensitivity (96% vs 76%), accuracy (94% vs 83%), and negative predictive value (95% vs 76%). CT was shown to be more accurate in staging periappendiceal inflammation, more useful in diagnosing acute abdominal conditions unrelated to appendicitis, and more sensitive in demonstrating a normal appendix and in excluding acute appendicitis from the differential diagnosis. Further prospective investigation is needed to confirm these results.
Use for provision of an alternate diagnosis to explain the patient's symptoms is a benefit of both US and CT in the patient suspected to have acute appendicitis (51,65,78,95). The differential diagnosis includes all surgical and nonsurgical conditions that cause abdominal pain, since appendicitis may mimic any of these diseases. The most common conditions discovered at false-negative appendectomy vary from study to study but include, in approximate order of frequency, abdominal pain of unknown cause, pelvic inflammatory disease and other acute gynecologic disorders, mesenteric lymphadenitis, acute gastroenteritis and other acute gastrointestinal tract diseases, and urinary tract infection and obstruction.
Careful US and CT evaluation of the right lower quadrant will result in a correct diagnosis in most patients. Although intestinal diseases often demonstrate overlapping imaging features, the addition of relevant clinical history usually narrows the differential to a few diagnostic entities. In certain gastrointestinal disorders, a constellation of findings may be noted that permits a precise diagnosis to be made.
A suggested approach is to first confirm or exclude the diagnosis of acute appendicitis. Once the appendiceal region is cleared, the cecum and ascending colon should be carefully examined for potential involvement by cecal neoplasm, cecal diverticulitis, typhlitis, or segmental colitis. Diseases that involve primarily the pericolonic fat, such as primary epiploic appendagitis and omental infarction, are then excluded.
Focus is then turned to the terminal ileum and its subtended mesentery. Gastrointestinal diseases to consider in this anatomic location include acute terminal ileitis, mesenteric lymphadenitis, and Crohn disease. Genitourinary disease then should be excluded, including acute pyelonephritis, ureteral obstruction, complications of ovarian cysts and masses, and acute postpartum ovarian vein thrombosis. In adult patients, one must also consider acute cholecystitis, which may mimic acute appendicitis if the enlarged gallbladder extends into the right-lower quadrant; pancreatitis; sigmoid diverticulitis; bowel ischemia; and bowel obstruction.
US and CT have replaced barium enema examination as the primary means of examining patients suspected to have appendicitis. Barium enema examination is not obsolete, however, and may be useful in evaluating complex colonic abnormalities detected with cross-sectional imaging (51). While investigators have explored the potential of magnetic resonance imaging (97) and radioactive isotope imaging (98,99) in patients with acute appendicitis, there is no current practical role for these imaging modalities in this patient population.
EFFECT OF IMAGING ON CLINICAL OUTCOMES
Outcome studies in which the medical and financial implications of radiologic imaging in patients suspected to have acute appendicitis are assessed have begun to appear in the literature. Recent investigations (55,59) have shown that the judicious use of CT in patients with equivocal clinical findings results in lower false-negative appendectomy rates when compared with the historical rate of 15%–20% promoted by aggressive surgical philosophy. Balthazar et al (55) demonstrated that CT led to an overall false-negative appendectomy rate of 4%, with a rate of 8.3% in female patients of childbearing age. This was accomplished without incurring an increase in the perforation rate, which, at 22%, was similar to that in previously published reports (8,12–14,26,27,35–42). Schuler et al (59) used CT to achieve a false-negative appendectomy rate of 6% (three of 52 appendectomies) in patients with clinically equivocal appendicitis. This figure was substantially lower than the 21% (11 of 52 appendectomies) false-negative appendectomy rate observed in a control group of patients who immediately underwent surgery after being judged clinically likely to have appendicitis.
Similar results have been reported by very experienced sonologists. Ooms et al (95) documented their experience with graded compression US over 4 years in patients who were clinically suspected to have acute appendicitis but were without signs of an appendiceal mass. The false-negative appendectomy rate at Ooms et al's institution decreased from 32% in 1985 to 12% in 1986 and to 7% from 1987 to 1989. At the same time, delay in treatment beyond 6 hours, due to diagnostic uncertainty in patients with surgically proved acute appendicitis, decreased from 11% to 4% to 2% over these same periods. These investigators noted that although US enhanced diagnostic accuracy, it could not entirely replace careful clinical evaluation and observation. This was emphasized by the fact that 16% of patients in this study who had normal or nondiagnostic US results eventually received a diagnosis of acute appendicitis.
Well-performed cost-effectiveness studies are needed to evaluate the utility of diagnostic imaging in this clinical setting. In the most detailed study to our knowledge to date, Rao et al (100) showed that the routine use of appendiceal CT in patients in the emergency department who meet clinical criteria for hospital admission for suspected appendicitis improves patient care both by averting unnecessary appendectomy and by averting delays before necessary medical or surgical treatment. Their cost analysis demonstrated that this imaging philosophy improved the use of hospital resources, because savings achieved by eliminating unnecessary surgery and in-hospital observation outweighed the cost of performing routine appendiceal CT.
The importance of avoiding unnecessary surgery was also stressed by Schuler et al (59), who evaluated the added expense of performing routine abdominal CT in patients suspected to have appendicitis and compared it with the expense of false-negative appendectomy. These investigators analyzed 1997 Medicare reimbursement rates for these procedures and found that if a hospital's false-negative appendectomy rate is 13%–15% or higher, it would be cost-effective to perform abdominal CT in every patient suspected to have appendicitis to achieve a 7% false-negative appendectomy rate. The authors noted that their study results were based on the use of Medicare payment data and that their conclusions may be limited because these figures did not accurately represent the true cost of the services provided.
LAPAROSCOPIC APPENDECTOMY
The surgical removal of the appendix prior to perforation is the goal of treatment in patients with acute appendicitis. Although appendectomy procedures are performed conventionally by using an open laparotomy technique, laparoscopic appendectomy has become an increasingly popular technique in recent years. First performed by Semm, a German gynecologist, in 1980 (101,102), laparoscopic appendectomy was met originally with disbelief in the surgical community, and it was not until 1988 that the first laparoscopic cholecystectomy was performed in the United States.
Although the current response to laparoscopic appendectomy has not been uniformly favorable (103–105), comparative studies (106) and meta-analyses of randomized controlled trials (107,108) suggest that laparoscopic appendectomy has several distinct advantages over the open technique, which include reduced length of hospital stay, faster return to normal activity, fewer wound infections, earlier return to consumption of solid foods, and some decreased demand for narcotic analgesia (109). The cosmetic benefit, with decreased scar production, is a strong motivation on the part of the patient in choosing between open and laparoscopic procedures (Poulin EC, oral communication, 1999). Increased duration of operating time and use of disposable products are associated with an increased overall cost for the laparoscopic technique as compared with that for open appendectomy (104). Nonetheless, this procedure has enthusiastic advocates and is now widely performed in spite of some continuing skepticism (Poulin EC, oral communication, 1999). The laparoscopic approach may not be optimal in patients with complications of appendicitis; in some centers, preoperative imaging is now being used to triage patients appropriately for laparoscopic versus open surgery.
SUMMARY
Helical CT and graded compression color Doppler US are powerful imaging tools that substantially improve diagnostic accuracy in patients with clinically equivocal appendicitis. Both imaging modalities have changed the approach to the diagnosis of acute appendicitis, although their exact roles as diagnostic aids are still being defined. Opinion now varies as to whether these tests should be performed in all patients suspected to have acute appendicitis or if they should be reserved for select patients with atypical or confusing clinical presentations.
Results of retrospective studies have shown that diagnostic imaging may improve patient care by lowering the false-negative appendectomy rate (55,59,95,100). There is a critical need to perform well-designed prospective studies to confirm this observation and to address other measures of clinical outcome and cost-effectiveness.
CT, US, and serial patient observations should be compared in a cost-benefit analysis to determine the value of appendiceal imaging in patients who are at intermediate risk for appendicitis and who typically are admitted for serial observation. Preadmission imaging in these patients may lead to earlier diagnosis, lower in-hospital perforation rates, and reduced hospital stays.
There is also a need to prospectively compare the accuracy of diagnostic imaging with that of surgical decision making in patients who have a high risk of appendicitis. These results are needed to define the potential role of appendiceal imaging in patients who normally undergo immediate surgical exploration. Because appendectomy costs up to 25 times as much as imaging, it may be cost-effective to perform imaging in most, if not all, patients.
Finally, future investigations are also needed to compare the diagnostic accuracy and cost-effectiveness of helical CT with those of graded compression US in both pediatric and adult patients. There is no question that diagnostic imaging has changed the "rules of the game"; now is the time to define these rules.
References
