Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Bacterial infection of the hepatic parenchyma frequently is recognized as multiple, small inflammatory foci (hepatic microabscesses) observed as an incidental finding in infants dying with sepsis. Diffuse hepatocellular damage, often in conjunction with infection of several organ systems, may be present after transplacental passage of microorganisms to the fetal circulation. Liver involvement rarely may take the form of a solitary purulent abscess. Metastatic focal infections of the liver associated with bacteremia resolve with antimicrobial therapy, are not recognized, or are found only at postmortem examination. Rarely, they are clinically apparent as solitary or multiple large abscesses diagnosed during life.
Although metastatic infections are rare, it is difficult to ascertain their true incidence. In a survey of more than 7500 autopsies of children performed from 1917 to 1967, Dehner and Kissane found only 3 neonates with multiple, small, pyogenic hepatic abscesses, whereas a review of approximately 4900 autopsies performed at Los Angeles Children’s Hospital from 1958 to 1978 revealed 9 infants with pyogenic hepatic abscesses. Among 175,000 neonates admitted from 1957 to 1977 to Milwaukee Children’s Hospital, 2 died with hepatic microabscesses ; 3 patients with hepatic microabscesses were seen among 83,000 pediatric patients admitted to New York Hospital from 1945 to 1983.
Solitary hepatic abscesses in newborns have also been reported rarely. These infections frequently are associated with prematurity and umbilical vein catheterization, or because of bacteremia, for instance, after sepsis caused by Staphylococcus aureus.
Etiologic agents in the infants described by Dehner and Kissane, Moss and Pysher, Chusid, and Simeunovic and colleagues have included Escherichia coli, S. aureus, Pseudomonas aeruginosa, Klebsiella spp., Enterobacter spp., Neisseria gonorrhoeae , and Listeria monocytogenes . The causative bacteria of solitary abscesses are generally the bacteria colonizing the umbilical stump, including S. aureus (most common), E. coli , Enterobacter spp., Klebsiella pneumoniae , P. aeruginosa , and Streptococcus pyogenes . The observation of gas in abscesses may indicate infection with anaerobes, a frequent cause of liver abscess in adults.
The most common cause of intrauterine bacterial hepatitis, congenital listeriosis, characteristically involves the liver and adrenals (see Chapter 13 ). Typical lesions are histologically sharply demarcated areas of necrosis (miliary granulomatosis) or microabscesses containing numerous pleomorphic gram-positive bacilli. Intrauterine tuberculosis results from maternal bacillemia with transplacental dissemination to the fetal bloodstream (see Chapter 18 ). Because the liver is perfused by blood and well oxygenated and is the first organ that encounters tubercle bacilli, it is often severely involved. The presence of primary liver foci is considered evidence for congenital tuberculous infection as a result of hematogenous spread through the umbilical vein. Closed-needle biopsy may be less accurate in the diagnosis of hepatic granulomas, and open biopsy may be required to confirm liver and regional node involvement. In addition to hepatomegaly, a clinical picture of fever with elevated serum immunoglobulin M (IgM) and chorioretinitis (e.g., choroid tubercles) may be similar to that caused by other congenital infectious agents. In a review by Abughal and coworkers, positive sites of culture for tuberculosis included liver (8 of 9), gastric aspirate (18 of 23), tracheal aspirate (7 of 7), ear (5 of 6), and cerebrospinal fluid (3 of 10).
Treponema pallidum is the spirochete most commonly associated with transplacental hepatic infection (see Chapter 16 ). Pathologic changes in the liver, which may be found in 95% of infants dying with congenital syphilis, include diffuse hepatitis or focal areas of inflammation, both frequently accompanied by increased connective tissue and enlargement of the liver. Involvement of liver has also been documented, on the basis of isolation of organisms or their identification in histologic sections, in newborns with intrauterine infection caused by various Leptospira spp. Transplacental infection of the fetus with Borrelia recurrentis causes little or no inflammation of liver parenchyma or biliary epithelium despite the presence of numerous spirochetes in the sinusoids.
With increasing use of broad-spectrum antibiotics, candidal sepsis has emerged as a serious complication in newborn intensive care units, and liver abscesses can develop during the course of disseminated infection. Use of umbilical venous catheters and total parenteral nutrition was reported in the majority of such cases.
Infectious agents may reach the liver of the fetus or newborn by one of several pathways: transplacental or transoral intrauterine infection; extension of thrombophlebitis of the umbilical vein, through the hepatic artery during the course of a systemic bacteremia; pyelophlebitis caused by a focus of infection in the drainage of the portal vein (mesenteric or splenic veins); direct invasion from contiguous structures or because of trauma or surgical inoculation; and extension up the biliary passages in cases of suppurative cholangitis. Abscesses with no apparent focus of infection seem to be common in newborns compared with older children. In one such case of solitary liver abscess, the nature of the lesion suggested that an umbilical vein infection, obscured by the large collection of purulent material in the abscess, was the probable pathogenesis.
The mode of infection usually determines the pattern of hepatic involvement. Intense and prolonged seeding of the liver parenchyma, such as that which occurs in conjunction with intrauterine infection or neonatal sepsis, almost invariably results in diffuse hepatocellular damage or multiple small inflammatory lesions. Umbilical vein thrombophlebitis may cause an abscess of the falciform ligament or extend into a single branch of the portal vein to produce a solitary pyogenic abscess, or it can lead to disseminated foci of infection through dislodgment of septic emboli.
The frequent use of umbilical catheters has been associated with an increase in the numbers of infants with solitary or multiple hepatic abscesses. In three large series, including almost 500 infants who died after placement of umbilical vein catheters, 29 infants were found to have purulent infections of hepatic vessels or parenchyma. Use of venous catheters for infusion of hypertonic or acidic solutions may provide a necrotic focus for abscess formation, and prolonged or repeated catheterization of a necrotic umbilical stump provides an ideal pathway for introduction of pathogenic organisms.
Although neonatal liver abscesses usually are caused by hematogenous dissemination of bacteria through the hepatic artery or umbilical vein, examples of infection arising from various other sources have been described. Solitary abscesses have followed a presumed portal vein bacteremia caused by amebic colitis. Direct invasion of adjacent liver parenchyma from purulent cholecystitis or postoperative perihepatic abscesses also has been observed. Ascending cholangitis, the most frequent cause of hepatic purulent infections in adults, has not been implicated in the causes of newborn infections.
Descriptions of “umbilical sepsis” and “acute interstitial hepatitis” recorded by Morison seem to indicate that his patients had acquired bacterial infections of umbilical vessels with widespread extension into portal tracts. Although mild periportal parenchymal necrosis was observed in a few infants, hepatocellular damage was minimal or absent in most. Similar lesions have been found in infants dying with sepsis and infantile diarrhea.
Multiple hepatic abscesses and diffuse hepatitis related to neonatal sepsis or transplacental fetal infection are usually recognized only at autopsy. Very few clinical manifestations referable to hepatocellular damage are evident before death. The signs and symptoms associated with these conditions are those of the underlying sepsis or of secondary metastatic complications, such as meningitis, pneumonitis, or peritonitis. Solitary abscesses are indolent in terms of their development and clinical presentation. Although the suppurative umbilical focus or umbilical catheterization responsible for the introduction of microorganisms can usually be traced to the first week of life, evidence of hepatic involvement is usually not apparent before the second or third week. The abscess frequently becomes a source for the hematogenous dissemination of microorganisms so that most infants have signs and symptoms of a bacteremia. Despite intense infection of the underlying vessels, the umbilical stump usually shows no evidence of inflammation or purulent discharge. The presence of hepatomegaly, a finding commonly associated with neonatal sepsis, also offers little aid in establishing a definitive diagnosis. In one half of infants for whom physical findings are clearly described, a well-delineated, often fluctuant or tender mass could be palpated in the epigastrium or right upper quadrant. On a few occasions, the infant’s mother noticed a mass, often several days before the onset of systemic symptoms. Abscesses occur in the right or left lobe of the liver with almost equal frequency and are generally 3 cm or greater in diameter at the time of surgical exploration.
Hematologic studies are of little value in establishing a diagnosis; leukocyte counts and sedimentation rates may be normal or elevated. The serum levels of liver enzymes may also be normal or elevated. Abdominal radiographs are usually normal or show nonspecific displacement of the lower edge of the liver. In five infants, diagnosis was suspected from plain radiographs by the presence of gas within the hepatic shadow. Radiologic findings that commonly accompany hepatic abscess in older children, such as an altered contour of the diaphragm, right pleural effusion, and platelike atelectasis, are rarely present in neonates.
Ultrasonography should be the initial imaging study in newborns with clinical evidence of a hepatic abscess. If ultrasonography is negative and the diagnosis is still strongly suspected, more sensitive techniques, such as computed tomography (CT) or magnetic resonance imaging (MRI), should be performed. Enhancement with contrast agents may increase the definition of smaller abscesses. Because congenital cysts, arteriovenous malformations, and tumors with central necrosis or hemorrhage can mimic hepatic abscess, the diagnosis should always be confirmed by aspiration of purulent material at laparotomy or by means of percutaneous drainage with ultrasound or CT guidance.
The prognosis for infants with diffuse liver involvement related to fetal or neonatal sepsis depends on the underlying condition because hepatic function is rarely compromised sufficiently to determine the outcome. In most cases, pathologic changes in the liver are unsuspected before postmortem examination.
Of 28 infants with solitary hepatic abscesses whose course was described, 6 died. Four newborns died with sepsis caused by organisms that were identical to the organisms isolated from the abscess, and the death of another was ascribed to cecal perforation. Prematurity was undoubtedly a major contributing factor in two of these deaths.
Newborns with a solitary hepatic abscess have traditionally been treated with open surgical drainage in conjunction with antibiotic therapy. Percutaneous catheter drainage is less invasive and often is the preferred first treatment. Several investigators have described the use of percutaneous drainage of intrahepatic abscesses and cysts, guided by CT or ultrasonography, in neonates. When combined with antibiotic therapy and monitored by ultrasonography to ensure resolution, this treatment has been highly effective. It is questionable whether drainage contributed to recovery other than by aiding the selection of antibiotic coverage. Subsequently, patients have been successfully treated with empirical antibiotic therapy alone.
The risk of bacteremia and disseminated infection is high in neonates, and the need to identify infecting organisms to guide antibiotic coverage is of greater urgency in the first weeks of life. It is appropriate to ascertain a microbiologic diagnosis with radiographically guided aspiration or drainage of hepatic abscess in a newborn. When proper equipment (e.g., CT, ultrasonography) and experienced personnel are available, this can be attempted percutaneously. When they are unavailable, open surgical drainage should be performed. Empirical antibiotic therapy should be reserved only for infants for whom it is believed that the risk of open or closed drainage would exceed the potential benefits.
If purulent material is obtained, initial antibiotic therapy can be selected on the basis of Gram stain. In addition to S. aureus and the aerobic enteric organisms commonly associated with hepatic abscesses, anaerobic bacteria have been suspected as the cause of infection in numerous patients. If foul-smelling pus is aspirated or if Gram-stained smears show organisms with the characteristic morphology of anaerobes, metronidazole, β-lactam and β-lactamase inhibitor combinations (e.g., piperacillin and tazobactam), clindamycin, or imipenem should be included in the initial regimen. Cultures of blood, cerebrospinal fluid, and urine should also be considered before initiation of therapy.
If empirical antibiotic therapy is required, it must be adequate for infections caused by S. aureus, enteric organisms, and anaerobic bacteria. The combination oxacillin, gentamicin, and clindamycin is appropriate. In nurseries where methicillin-resistant S. aureus (MRSA) or methicillin-resistant Staphylococcus epidermidis infections have been a problem, substitution of vancomycin for oxacillin can provide coverage for these organisms. Gentamicin (and other aminoglycosides) and vancomycin levels must be monitored and dosages adjusted as necessary. Extended-spectrum cephalosporins (e.g., cefotaxime, cefepime, ceftazidime) and carbapenems (e.g., meropenem) may be used for enteric organisms and Pseudomonas spp., often obviating the need for aminoglycosides. β-Lactam and β-lactamase inhibitor combination drugs (e.g., piperacillin and tazobactam or ampicillin and sulbactam) may provide coverage for many enteric organisms and anaerobic bacteria.
Definitive therapy is based on results of bacteriologic cultures that identify the bacteria and its antibiotic susceptibility. Adequate anaerobic transport and culture techniques must be available if meaningful information is to be obtained. Duration of treatment is based on clinical response, cessation of drainage, and resolution of the abscess cavity as determined by serial ultrasound examinations. Parenteral therapy should be maintained for at least 2 weeks, and longer-term therapy may be administered when necessary. In older children with multiple abscesses or in children for whom surgery is not feasible, therapy for 6 weeks or more has been recommended.
Similar to hepatic abscesses, splenic abscesses have been rarely described in infants. Only 1 of 55 splenic abscesses occurred in an infant younger than 6 months. S. aureus, Candida spp., and streptococci were the most frequent causes. In 20 of 48 cases, hepatic abscesses coexisted with splenic abscess. In the single infant case, torsion of the splenic vessels was present, whereas in older children, other distant infections of hematologic conditions (e.g., hemoglobinopathy, hematogenous malignancy) were the associated comorbid conditions. Rare etiologies of splenic abscess reported in the neonatal period include tuberculosis, with accompanying thrombocytopenia, and Entamoeba histolytica , complicated by fatal colonic perforation.
The development of ultrasonography has provided a safe and rapid means for evaluating the neonatal biliary tract. Consequently, an increasing number of reports have appeared describing ultrasound changes seen in the first month of life, with hydrops, cholelithiasis, and transient distention of the gallbladder associated or unassociated with sepsis. Ultrasound criteria for separating normal from pathologically enlarged gallbladders and biliary tracts in neonates have also been described.
Despite advanced technology and increased surveillance, cholecystitis in the neonate is observed infrequently. The literature has documented a few dozen cases, of which 9 were seen in association with an epidemic of neonatal enteritis caused by Salmonella enteritidis. Two cases of acute necrotizing cholecystitis caused by E. coli infection were described in a preterm and term neonate. The pathogenesis of this condition is uncertain; and the great majority of cases in the newborn period have been acalculous. It is postulated that sepsis, dehydration, prolonged fasting (e.g., total parenteral nutrition), congenital obstruction, or a stone impacted in the cystic duct leads to biliary stasis and acute distention of the gallbladder. In most cases, resolution of the primary process permits restoration of the flow of bile and relief of distention. In some cases, prolonged obstruction leads to hydrops. Cholecystitis rarely follows, perhaps because of a direct toxic effect of retained bile or because of ischemia related to elevated intraluminal pressure. Bacterial invasion by fecal flora is probably a secondary phenomenon. Other organisms that have been isolated from gallbladder contents or tissue include Serratia marcescens, Pseudomonas spp., Enterococcus faecalis, viridans streptococci, S. aureus, and Clostridium welchii .
Infants with cholecystitis may become ill at any time during the first weeks of life; most cases are diagnosed in the third or fourth week. The typical clinical picture is one of sepsis together with signs of peritoneal inflammation and a palpable tender right upper quadrant or epigastric mass. Diarrhea frequently accompanies these findings. Although ultrasonography and radionuclide scintigraphy are helpful in suggesting the presence of gallbladder enlargement or inflammation, diagnosis can be confirmed only by surgical exploration. Treatment consists of cholecystectomy or tube cholecystotomy combined with systemic antimicrobial therapy based on Gram stain, culture, and susceptibility studies. If a T tube is placed in the gallbladder, a cholangiogram should be obtained to confirm patency of the biliary system before the tube is removed.
Changes compatible with a diagnosis of ascending cholangitis have been described in histologic sections of liver specimens from infants who died with diarrhea accompanied by hepatocellular injury with cholestasis. Bacteria were also identified in the biliary tree of 2 of 178 premature infants who died after placement of an umbilical venous catheter for an exchange transfusion or for delivery of parenteral fluids. The reasons for this association, if any, are unclear. An infant with spontaneous cholangitis caused by Enterobacter agglomerans, presenting as a fever of unknown origin at 3 weeks of age, has also been reported. Severe inflammation and fibrosis of extrahepatic bile ducts and diffuse changes in the portal tracts, resembling changes found in biliary atresia, were found in a premature infant who died 3 hours after birth of listeriosis. The investigator postulated that occult prenatal infections with L. monocytogenes might be a rare cause of ascending cholangitis manifesting as idiopathic biliary atresia at birth.
Multiple adrenal microabscesses are occasionally found as metastatic lesions associated with neonatal sepsis. These abscesses are particularly characteristic of neonatal listeriosis (see Chapter 13 ). Solitary adrenal abscesses are rare, however; only a few dozen such cases have been described. The spectrum of organisms responsible for adrenal abscesses is the same as that seen in neonatal sepsis and includes E. coli, group B streptococci (GBS), Proteus mirabilis, S. aureus, Bacteroides spp., and Peptostreptococcus.
About three fourths of neonatal cases are male, and adrenal abscesses are more commonly located on the right side (approximately two thirds of patients); bilateral lesions are uncommon. The same sex and laterality predominance are seen with adrenal hemorrhage in the newborn, and it has been postulated that formation of an adrenal abscess requires a preexisting hematoma as a nidus for bacterial seeding. This theory of pathogenesis is supported further by clinical observations and by objective evidence (e.g., curvilinear calcifications) documenting the presence of hemorrhage before development of an abscess.
Most infants with adrenal abscess have presented in the third or fourth week of life with signs of sepsis and an abdominal or flank mass. A history of difficult delivery or intrapartum asphyxia was observed in about one half of these infants, and significant maternal fever or infection during labor was observed in about one fourth. Although a few infants are afebrile when first evaluated, a palpable mass is almost always present. Abscesses are usually 6 to 8 cm in diameter, with some containing 200 mL of pus and measuring 12 cm in diameter or crossing the midline.
Laboratory studies are helpful in the evaluation of a possible adrenal abscess. Most infants exhibit a leukocytosis; about one third are anemic and with a history of prolonged neonatal jaundice, both of which are features associated with adrenal hemorrhage. Urinary excretion of catecholamines and their metabolites (particularly vanillylmandelic acid and homovanillic acid), which is usually increased with neuroblastoma, is normal. Because most infants with adrenal abscess are seen for evaluation of possible sepsis, a blood culture, lumbar puncture, urine culture, and chest radiograph should be obtained.
Ultrasonography has become a widely accepted modality for initial evaluation of all neonatal abdominal masses. With the presence of an adrenal abscess, ultrasound examination can help to define the extent and cystic nature of the lesion and often can show movable necrotic debris in the abscess cavity.
With serial examinations, abscesses can be distinguished from masses associated with liquefying hematoma, adrenal cyst, hydronephrosis of an obstructed upper pole duplication, or necrotic neuroblastoma. Intravenous pyelography shows downward displacement of the kidney and compression of the upper calyces, which confirms the presence of a suprarenal mass. A round, suprarenal, radiopaque halo or rim with central lucency, which is characteristic of adrenal abscess, may also be seen on early films but is not pathognomonic. Intravenous pyelography adds little diagnostic information to that provided by ultrasound studies. Experience with radionuclide scanning, CT, and MRI in this condition is limited, but these modalities are likely to be as useful as ultrasonography.
Whatever diagnostic methods are used, concern about persisting signs of sepsis and the possible presence of an adrenal neoplasm usually encourage early efforts to establish a diagnosis. In the past, recommended management has been incision and drainage or resection of the abscess. Needle aspiration under ultrasound guidance, combined with placement of a catheter for drainage and irrigation, has proved to be a useful alternative method and is likely to supplant open drainage as the preferred method. Antibiotic therapy should be based on Gram stain, culture, and susceptibility studies of abscess fluid and should be continued for 10 to 14 days, provided that drainage can be established.
The adrenals are infected in about 15% of infants with congenital syphilis. In addition to the presence of spirochetes, the most frequent and characteristic change is an extraordinary amount of cellular connective tissue in the capsule.
Acute appendicitis is extremely rare in infants younger than 4 weeks of age. Reviews of more than 25,000 cases of appendicitis in infants and children in Great Britain, Ireland, Germany, and the United States revealed only 8 infants who presented during the neonatal period. Since the condition was first described by Albrecht in 1905 and Diess in 1908, sufficient cases of neonatal suppurative appendicitis have been reported in the literature details that permit characterization of the clinical features. Infants with appendicitis caused by other conditions, such as Hirschsprung disease, necrotizing enterocolitis (NEC), or incarceration in an inguinal hernia, have not been included in this discussion.
Inflammation of the appendix is more common in newborn boys than newborn girls. In reports in which the sex was stated, 40 cases occurred in boys, and 17 cases occurred in girls. Prematurity also seems to be a predisposing factor: 23 of the 49 infants whose birth weights were recorded weighed less than 2500 g at birth. The incidence of appendicitis in infants of multiple births (six twins and one triplet) seems to be higher than would be expected on the basis of low birth weight alone.
Although rare, neonatal appendicitis is likely to occur with greater frequency, as perinatal conditions associated with appendicitis, such as prematurity and survival after severe hypoxia, are increasingly common with advances in neonatal care.
Because obstruction of the appendiceal lumen is responsible for almost all cases of appendicitis, it is intuitive that gram-negative enteric organisms resident in the bowel are usually isolated from the peritoneal fluid or periappendiceal pus of about 75% of infants. Specific etiologic agents include E. coli, Klebsiella spp., Enterobacter spp., Pseudomonas spp., Proteus spp., untyped streptococci, S. aureus, and Bacteroides spp. These bacterial species have also been isolated from the peritoneal fluid of older children with appendicitis. Attempts at isolation of anaerobic bacteria have been rarely described.
A case of perforated amebic appendicitis with secondary bacterial peritonitis and multiple hepatic abscesses in a premature infant born in Great Britain has been reported. E. histolytica observed in the wall of the necrotic appendix was presumably acquired from the infant’s father, who was a carrier. A patient with gangrenous appendicitis associated with Rhizopus oryzae has also been reported. It was postulated that the fungus colonized the infant’s gut by transfer from an adhesive bandage used to secure an endotracheal tube.
Obstruction of the appendiceal lumen has been generally accepted as the primary cause of appendicitis in all age groups. The relative rarity of this condition in the first month of life is probably related to factors that serve to decrease the likelihood of obstruction, including a wide-based, funnel-shaped appendix; the predominantly liquid and soft solid diet given to infants; the absence of prolonged periods in the upright position; and the infrequency of infections that cause hyperplasia of the appendiceal lymphoid tissue. The causes of luminal obstruction in the newborn period, when recognized, are often extrinsic to the appendix itself. Reports of appendicitis caused by the presence of ectopic pancreatic tissue, a fecalith, or meconium plug are unusual exceptions.
Inflammation of the appendix with perforation has been described as the presenting illness in several infants with neonatal Hirschsprung disease. The association of these two conditions has been attributed to functional obstruction, increased intraluminal pressure, and fecal trapping that occur proximal to aganglionic segments. Suppurative appendicitis related to incarceration and strangulation of the cecum within an inguinal or scrotal hernia has been found in numerous infants.
The onset of neonatal appendicitis generally occurs during the first 2 weeks of life. Only 3 of 54 infants with this condition presented between the 1st and 10th day. The reasons for this phenomenon are unclear, particularly in view of the relatively even distribution of cases during the remainder of the first year of life. At least five cases of “prenatal” appendicitis have been described. Of the four available for analysis, only one showed definite evidence of a suppurative process in the appendix and signs of bowel obstruction clearly present at birth ; however, cultures and Gram stain of the pus found at surgery were free of bacteria. Poisoning by mercuric chloride was suspected in one of the remaining three cases, and the other two, who were said to have prenatal rupture of the appendix, were asymptomatic until the 2nd and 12th days of life.
The signs of neonatal appendicitis correspond to the signs of any of the various forms of intestinal obstruction that occur during the newborn period ( Table 10-1 ). Prominent early findings include abdominal distention; progressive and frequently bilious vomiting; and evidence of pain, as manifested by persistent crying, irritability, or “colic.” Clinical features such as diarrhea, constipation, lethargy, or refusal to feed may also be evident but are too nonspecific to be helpful in establishing a diagnosis. The presence or absence of fever is an unreliable sign in appendicitis as in other forms of neonatal infection; temperature has been recorded as normal or subnormal in greater than 50% of newborns with this condition. Abdominal tenderness and guarding are inconsistent findings and, when present, are rarely localized to the appendiceal area. Physical signs of sufficient specificity to indicate acute inflammation of the appendix are generally absent until late in the course of the illness, when gangrene and rupture may result in the formation of a localized intraabdominal abscess or cellulitis of the anterior abdominal wall. Erythema or edema, or both, of the right lower quadrant has been observed in several patients. The presence of this finding, particularly when accompanied by a palpable mass in the right iliac fossa, indicates bowel perforation with peritonitis and should suggest a preoperative diagnosis of NEC or appendicitis (see “Necrotizing Enterocolitis” ).
Sign | Incidence (%) |
---|---|
Abdominal distention | 90 |
Vomiting | 60 |
Refusal of feedings | 40 |
Temperature ≥38° C | 40 |
Temperature 37-38° C | 30 |
Temperature ≤37° C | 30 |
Pain (crying, restlessness) | 30 |
Lethargy | 30 |
Erythema/edema of right lower quadrant | 25 |
Mass in right lower quadrant | 20 |
Diarrhea | 20 |
Passage of bloody stools | 20 |
The diagnosis of appendicitis in a neonate is usually determined at surgery performed for evaluation of abdominal distention and suspected peritonitis. With the high incidence of prematurity associated with early appendicitis, bowel perforation from NEC has been a common preoperative consideration. The two conditions can coexist, and in some cases, the appendix may participate in the process of ischemic necrosis and perforation.
Laboratory studies are of little value in establishing a diagnosis of appendicitis in a newborn. White blood cell counts of less than 10,000/mm 3 were found in 10 of 30 infants. Urinalyses are usually normal, although ketonuria, which reflects diminished caloric intake, and hematuria and proteinuria may be seen. Because bacteremia may accompany appendiceal perforation and peritonitis, a blood culture and evaluation for metastatic infection with lumbar puncture and chest radiography should be performed. The value of paracentesis for diagnosis of bowel perforation and peritoneal infection is discussed later (see “Necrotizing Enterocolitis” ).
Radiologic examinations are occasionally helpful but, in most cases, serve only to confirm a clinical impression of small bowel obstruction. The presence of an increased soft tissue density displacing loops of intestine from the right iliac fossa generally indicates appendiceal perforation with abscess formation and is perhaps the most reliable sign of acute appendicitis in the neonate. Extraluminal gas may be localized briefly to the right lower quadrant after rupture of the appendix. The rapid development of an extensive pneumoperitoneum obscures the site of origin of the escaping gas in most infants within a short time. Ultrasonography may aid in detection of a periappendiceal abscess but is not helpful in establishing an early diagnosis of appendicitis because it lacks sensitivity and specificity.
The overall mortality rate from appendicitis in the newborn is high but is improving. Eight of the newborns in the last 12 reported cases have survived, whereas of 60 infants with this condition for whom the outcome was recorded, 38 (64%) died. Survival was unrelated to birth weight. Among factors responsible for mortalities, three seem to be of primary importance: delay in diagnosis, a high incidence of perforation, and the rapid onset of diffuse peritonitis after appendiceal rupture.
Perforation has been identified at surgery or autopsy in 70% of newborns with acute appendicitis. The relative frequency of this complication has been attributed to delays in establishing a diagnosis and to certain anatomic features of the appendix in young infants that predispose it to early necrosis and rupture. These features include a meager blood supply that renders the organ more vulnerable to ischemia; a cecum that is relatively smaller and less distensible than that of adults, forcing a greater intraluminal pressure on the appendix; and the presence of a thin muscularis and serosa that readily lose their structural integrity under the combined effects of ischemia and increased internal pressure.
After the appendix ruptures, infants are unable to contain infection efficiently at the site of origin. Rapid dissemination of spilled intestinal contents produces a diffuse peritonitis within hours because of the small size of the infant’s omentum, which fails to provide an efficient envelope for escaping material; the relatively longer and more mobile mesenteries, which favor widespread contamination; and the small size of the peritoneal cavity, which also permits access of infected material to areas distant from the site of perforation. Peritonitis, accompanied by sepsis and by the massive outpouring of fluids, electrolytes, and proteins from inflamed serosal surfaces, is generally the terminal event in neonatal appendicitis. Deterioration of the infant’s condition is often extremely rapid; failure to recognize the underlying illness and to institute appropriate therapy promptly is inevitably followed by a fatal outcome.
Become a Clinical Tree membership for Full access and enjoy Unlimited articles
If you are a member. Log in here