Abstract
A 2-year-old girl with a past medical history of cutaneous mastocytosis and eczema presented with 1 day of yellow-green, nonbloody vomiting, bradycardia, and listlessness. She was evaluated by her pediatrician and sent to the emergency department because of concern for dehydration. In the emergency department, she improved with fluid rehydration but still had decreased energy and bradycardia. Her electrocardiogram revealed sinus bradycardia, and laboratory results did not reveal any electrolyte abnormalities. Glucose levels were normal. An abdominal radiograph revealed a moderate-to-large stool burden, and the results of a computed tomography scan of the head were normal. An abdominal ultrasound was obtained to evaluate for intussusception. The ultrasound revealed a blind-ending tubular structure in the right-lower quadrant with adjacent free fluid, which was concerning for appendicitis. The patient was admitted to the surgical service for further management and was taken to the operating room, where a definitive diagnosis was made.
- CT —
- computed tomography
- ECG —
- electrocardiogram
- ED —
- emergency department
- NAT —
- nonaccidental trauma
- RLQ —
- right-lower quadrant
Case History With Subspecialty Input
Dr Edmunds, Pediatric Emergency Medicine Fellow, Moderator
Chief Complaint
The patient presented with isolated vomiting, listlessness, and bradycardia.
History of Present Illness
A 2-year-old girl initially presented to her pediatrician’s office with vomiting. The patient had vomited 4 times that day and was unable to tolerate fluids. The first 2 episodes of emesis were described as yellowish green and nonbloody. Subsequent vomitus was nonbilious and looked like what she consumed. The family denied any fever or diarrhea. She had received no medications at home, and she had no complaints of abdominal pain. She was sent to the emergency department (ED) because of concern for dehydration.
Past Medical History
The patient had a past medical history of cutaneous mastocytosis and eczema, which was diagnosed at 7 months of age. She had 1 previous hospital admission for dehydration secondary to presumed viral gastroenteritis.
Physical Examination
The patient was afebrile with a heart rate of 82 beats per minute, a respiratory rate of 30 breaths per minute, a blood pressure of 125/55 mm Hg, and an oxygen saturation of 100% on room air. Her weight was 12.3 kg (39th percentile), and her height was 89 cm (55th percentile). The patient was drowsy but arousable, listless, and appeared clinically dehydrated. Her pupils were equal, round, and reactive to light. The mucus membranes were tacky. The neck was supple. The lungs were clear to auscultation bilaterally. Cardiac examination revealed a regular rhythm with a heart rate of 80 to 90 beats per minute. The abdomen was soft without discernible masses, guarding, or rebound. She did not cry or complain with palpation of her abdomen. The genitalia were normal. She had brisk central pulses and a capillary refill time of 3 seconds. She had no cranial nerve deficits, was oriented to her mother, and had no ataxia or deficits in coordination.
Laboratory Values
Her complete blood count was significant for a white blood cell count of 13.8 K/μL (normal: 6.0–17.0 K/μL), hemoglobin levels of 12.6 g/dL (normal: 11.5–13.5 g/dL), and a platelet count of 488 K/μL (elevated: 135–466 K/μL). Of note, she had a normal platelet count of 311 K/μL 1 week before. Her glucose and electrolyte levels revealed no abnormalities, and she had normal amylase and lipase levels. Her urinalysis had 20 ketones with a specific gravity of 1.031 and 30 mg/dL of protein.
Electrocardiogram
An electrocardiogram (ECG) was performed, which revealed a ventricular rate of 80 contractions per minute with no other abnormalities (Fig 1).
Patient’s ECG revealing sinus bradycardia.
Radiographic Imaging
An abdominal radiograph was performed, which revealed nonobstructive gaseous distension of the bowel with a moderate-to-large stool load. A computed tomography (CT) scan of the head without contrast was performed, which had normal results. An ultrasound for intussusception was performed. The ultrasound revealed no intussusception, but it revealed a blind-ending tubular structure in the right-lower quadrant (RLQ) with adjacent free fluid (Fig 2).
Blind-ending tubular structure in the RLQ with adjacent free fluid. A, Blind-ending tubular structure with hypoechoic wall without hyperemia. B, Free fluid surrounding tubular structure.
ED and Hospital Course
On arrival, given the clinical dehydration and listlessness, a peripheral intravenous drip was placed, laboratory tests were obtained, and fluid resuscitation was started expediently. Although she did show some clinical improvement, the results of laboratory tests and plain radiographs were not diagnostic. Emesis in the ED was nonbloody and nonbilious, and her abdomen remained soft and nonperitoneal on examination. While continuing fluid hydration, an ultrasound to evaluate for intussusception was performed, which suggested possible appendicitis. Surgery providers evaluated and admitted the patient, feeling her history and examination results were not consistent with a surgical process. While admitted, she developed more pronounced abdominal tenderness, and a repeat ultrasound was increasingly concerning for appendicitis. She was taken to the operating room, where a diagnostic laparoscopy revealed the cause of her symptoms (Fig 3).
Intraoperative images. A, Normal appendix. B, Torsed and hemorrhagic infundibulopelvic ligament and ovary. C, Torsed and hemorrhagic infundibulopelvic ligament and ovary. D, Detorsed pedicle.
Dr Zamor, as a general pediatrician who might see this child in an office, what is your differential diagnosis? Would you consider doing the evaluation with the child as an outpatient or refer her to the ED?
Dr Zamor, General Pediatrician
Isolated vomiting has an extensive differential diagnosis. Narrowing this differential is guided by the child’s vital signs and physical examination findings. In a previously healthy and well-appearing child, the most common causes are infectious, with viral causes being more common than bacterial. Other common infectious causes of vomiting in a young pediatric patient are streptococcal pharyngitis or a urinary tract infection, particularly in girls. Electrolyte abnormalities, including disorders of glucose, are also associated with vomiting, either as a primary or secondary cause. In this child with acute vomiting without diarrhea, I would also consider anatomic causes, especially if the child has abdominal tenderness on examination. These include surgical emergencies, such as malrotation with volvulus (although patients often present in infancy with persistent bilious emesis, they can present later in childhood), appendicitis, pelvic pathology (although this is rare in a premenarchal child), intussusception, or severe constipation with impaction. Because this child had relative bradycardia and was sleepier than I would expect for her history, I would be concerned about an intracranial process (such as a brain tumor), especially if the vomiting was associated with headache, positional changes, early-morning vomiting, or abnormalities in any part of the neurologic examination. Unfortunately, nonaccidental trauma (NAT) is also always a possibility in pediatrics. NAT can include head bleeds, abdominal trauma, or cardiac contusions.
Deciding whether to refer a child to the ED should be based on clinical presentation. If the patient has normal vital signs and is well appearing on examination, a point-of-care glucose test to evaluate for hyper- or hypoglycemia could be obtained followed by an attempt at oral rehydration (often after a trial of oral antiemetic). However, given this patient’s drowsy appearance and abnormal physical examination, I would be concerned for significant dehydration and possibly a more noteworthy underlying cause. I would refer her to the ED to expedite resuscitation and evaluation.
Dr Edmunds
Dr Hariharan, what was your differential diagnosis after evaluating this patient?
Dr Hariharan, Pediatric Emergency Medicine
The most concerning part of the examination was the relative bradycardia combined with listlessness that were out of proportion with her clinical examination findings (minimal abdominal tenderness, minimally tacky mucus membranes, brisk capillary refill, and a normal cardiac examination aside from the bradycardia). I initially thought this child had dehydration, possibly with an electrolyte abnormality. The decreased energy could be explained by hypoglycemia. Hyperglycemia can cause isolated vomiting with mental status changes. Hyponatremic dehydration can also present with listlessness without clinically apparent dehydration. Potassium abnormalities can present with cardiac changes. Once the electrolyte levels were confirmed as normal, I had to broaden my differential diagnosis. My next consideration was infection. However, she did not have a fever, and she had a supple neck without any evidence of meningismus; her urine did not suggest a urinary tract infection or pyelonephritis; she had no diarrhea to attribute the vomiting to uncomplicated gastroenteritis; and an ECG revealed isolated bradycardia without cardiac interval or voltage changes to suggest cardiac infection or inflammation. As a result, with isolated vomiting and an abnormal mental status, I expanded the differential further to include central nervous system lesions, specifically a mass or a bleed. At her age, NAT with intraabdominal injury or intracranial injury were also strongly considered. With a normal head CT scan result, my focus shifted to her abdominal examination. Her mother stated she had not complained of any pain, and she did not have any guarding or rebound on exam. I could convince myself that with deep palpation of the RLQ, she reacted a little more than she did on the examination of any other quadrant. She did not have any change in her lipase or amylase levels to suggest pancreatitis. Although malrotation was a consideration, especially because she had been admitted once before for a significant gastrointestinal illness, intussusception seemed more likely given her age, mental status changes, and possible RLQ pain. Although malrotation is certainly a more catastrophic abdominal diagnosis, and although her initial 2 episodes of emesis had been yellow-green, her subsequent vomit was nonbilious, she was older than expected for malrotation, and the inconsistent abdominal pain I could elicit was focused on her right side; therefore, I opted to preferentially evaluate for intussusception. Given that toddlers with appendicitis present with atypical symptoms, I considered appendicitis, but without fever and the abrupt onset of symptoms, this seemed less likely. Finally, I considered an ingestion (although the mother denied any medications or risk of ingestion) or a malignancy.
Dr Edmunds
Dr Shah, could this patient’s presentation have been an exacerbation of her mastocytosis?
Dr Shah, Pediatric Dermatology
Cutaneous mastocytosis involves only the skin and usually does not result in systemic manifestations without other associated symptoms. It is usually associated with tachycardia and not bradycardia.
Dr Edmunds
Dr Koberlein, while reviewing the ultrasound images for this patient, what was your differential? Are there any other studies you recommended performing to help determine the etiology?
Dr Koberlein, Pediatric Radiology
At her age, this patient bridged a few classic differentials,1 of which initiated the first imaging study. A single-quadrant ultrasound with attention to the colon (and additional limited 4-quadrant images) was obtained to evaluate for intussusception. The study failed to reveal any form of intussusception; therefore, the first major differential was excluded. In the absence of finding an intussusception, gastroenteritis was considered. On an ultrasound, gastroenteritis may present as hyperperistaltic bowel loops, sometimes with edema or adjacent free fluid. These findings were not identified on the ultrasound, thus decreasing the likelihood of this diagnosis. Mesenteric adenitis (often a diagnosis of exclusion) was considered, with borderline to mildly enlarged or clustered mesenteric lymph nodes usually being identified. Again, this was not seen. An additional consideration at this age would be a Meckel diverticulum, although this is not a classic presentation (usually painless rectal bleeding), nor was it consistent with the initial ultrasound findings. Additional differentials, such as inflammatory bowel disease, neoplasms, and lymphoma, may have been considered but did not match the initial ultrasound findings.
Of note, the study ordered was not intended to be used to assess for appendicitis or pelvic pathology in this female patient. However, many pediatric sonographers do a brief scan to assess for any local pathology or inflammation, and in this case, there was a significant finding in the RLQ. With the finding of a blind-ending tubular structure in the RLQ with adjacent free fluid, the most likely diagnosis would be appendicitis. It should be noted that appendicitis at <2 years of age is not common, and children <5 years of age frequently present with atypical histories or late in the course with perforation.1 Given her sex, pelvic pathology (including ovarian torsion) would be on the differential; however, given its rarity in this age group, we did not pursue a dedicated pelvic ultrasound.
A CT scan of the abdomen and/or pelvis could be considered if the examination result was indeterminate or if the clinical presentation and examination findings were not consistent with the ultrasound findings. Alternatively, MRI could also be considered, but at this patient’s age, a complete abdominal and/or pelvic examination would be challenging without sedation.
Dr Edmunds
The pediatric surgical service admitted the patient for serial abdominal examinations. After an enema for presumed constipation, she had increasing abdominal pain overnight. Therefore, a repeat ultrasound was performed in the morning, with findings reported as a noncompressible appendix with a diameter of 6 to 7 mm and tenderness in the RLQ with compression. There was moderate periappendiceal fluid in the abdomen, which appeared to have increased from previous ultrasound findings.
Dr Gurria, given this patient’s age as well as the ultrasound findings, how convinced are you that this represents appendicitis? What is the best next diagnostic step?
Dr Gurria, Pediatric General Surgery
The clinical picture of this patient was challenging. Her symptomatology, workup, and physical examination did not correlate with any specific diagnosis. Appendicitis was not the initial diagnosis that came to mind on assessing this patient given her age and symptoms. She presented with 2 episodes of nonbloody, yellowish-green emesis that raised a question of bilious emesis, but it transitioned into nonbilious emesis, and findings of her abdominal examination were benign with minimal discomfort. Given her age and her examination findings, the source of her symptoms was felt to be due to gastroenteritis, mesenteric adenitis, or a process outside of the abdomen causing generalized illness. The fact that she presented with moderate-to-severe constipation on plain radiographs, which could have also been the sole cause of her symptomatology, was just another cloud in her clinical picture. Lower on the differential was intussusception (which was ruled out with the ultrasound), ovarian pathology (given the ultrasound findings of free fluid and no clear findings of intussusception or appendicitis), Meckel diverticulitis, and finally appendicitis. Given her clinical picture, physical findings, and the episodes of nonbilious emesis, malrotation was not considered to be the cause of her condition. Both Meckel diverticulitis and appendicitis that present in this age group tend to cause significant morbidity given that they present late in the disease process, either with peritonitis from a perforation or with a septic picture. The initial ultrasound findings for our patient were nonspecific; however, with her progressively increased abdominal discomfort and the findings on the follow-up sonogram after bowel decompression with an enema, appendicitis was felt to be a possible cause. Although we were not totally convinced of the diagnosis, on the basis of her stagnant clinical status, increased abdominal pain, and concern for intraabdominal pathology, we decided that the next best diagnostic step was a diagnostic laparoscopy. Clinical history and physical examination results concerning for a surgical pathology that cannot be demonstrated in ancillary tests and diagnostic imaging should be followed with a surgical exploration. In this case, after a thorough multidisciplinary evaluation, the best next step was to perform surgery. Under these circumstances, diagnostic laparoscopy is an extremely valuable tool that can be used to confirm a clinical suspicion of an acute surgical problem while avoiding the potential associated morbidity of an open surgical exploration (if providers are able to continue laparoscopically or if no further intervention is needed).
Dr Edmunds
The patient was taken to the operating room that afternoon for a diagnostic laparoscopy. During laparoscopy, the appendix appeared normal, as did the left ovary and uterus. The right ovary was torsed and appeared edematous and hemorrhagic. The infundibulopelvic ligament was dark and hemorrhagic appearing (Fig 3). Under laparoscopy, the ovary was detorsed with visible improvement in ovarian perfusion. Gynecology experts were consulted intraoperatively. After discussion between the surgical service providers and the family, it was decided to leave the ovary in place rather than perform an oophorectomy. The patient has been managed by gynecology providers as an outpatient. A repeat ultrasound 3 months later revealed a normal right ovary, and the remainder of the pelvis was normal.
Discussion
Dr Edmunds
Dr Breech, how unusual is it to find ovarian torsion in a 2-year-old girl, and what are the most likely causes of torsion in this age group?
Dr Breech, Pediatric Gynecology
Although uncommon, ovarian torsion occurs in young children, including infants and newborns. Torsion has an estimated incidence of ∼4.9 in 100 000 children, as is noted in an analysis of the 2006 Kids’ Inpatient Database.2 In pediatric patients, 16% to 49% of torsions occur in normal ovaries, and the remaining episodes are associated with the presence of masses.3,4 In the absence of a mass, the cause of torsion in this age group remains less clear, with some authors hypothesizing anatomic variants, including hypermobility of an elongated ovarian ligament or an excessively lax mesosalpinx or mesovarium as contributing factors.5 With the more abdominal location of the ovaries in prepubertal girls, most girls will experience acute abdominal pain, often with fever, nausea, vomiting, and an elevated white blood cell count.6,7
Dr Edmunds
How often should patients be managed after a diagnosis of ovarian torsion?
Dr Breech
In the past, an oophorectomy was performed rather than detorsion of the ovary, which minimized the role of follow-up imaging. However, more recent literature supports the important role of detorsion and preservation of the reproductive structures.8,9 Although traditional teaching was that the time to diagnosis decreased the likelihood of ovarian salvaging, a study in 2005 suggested that the time to diagnosis did not affect ovarian salvaging,4 which further supports attempting detorsion. With current practice being to retain the detorsed adnexal structures, postsurgical imaging is important in ruling out a persistent mass that may have caused the torsion. However, to date, there are no guidelines for the timing of follow-up ultrasonography. Approximately 3 months after surgical detorsion, the ovarian edema should be decreased, allowing for an acceptable view of the ovarian stroma that can be used to rule out a persistent lesion, such as a cyst or mass.
Dr Edmunds
Dr Gurria, how common is it that an ultrasound is highly suspicious for appendicitis and the patient is later found to have a different underlying cause for his or her symptoms?
Dr Gurria
Ultrasonography is a good diagnostic tool for pediatric patients, especially in this patient’s age group and with her body habitus. It has a good sensitivity and specificity but should always be correlated with clinical findings. When an ultrasound suggests acute appendicitis in our institution, the chances of finding a normal appendix are extremely rare. When the diagnosis via imaging reveals that the likelihood of appendicitis is intermediate or that the appendix is not fully visualized, there is a chance (although it is low) of having a process other than appendicitis as the cause of RLQ pain, such as tubo-ovarian pathology. In this case, the clinical examination should guide the decision for a surgical intervention. When discordance exists between ultrasound results and clinical findings, frequent reassessments with serial abdominal examinations, additional diagnostic tests, or surgical exploration are paramount. This is true particularly in female patients, for whom the causes of RLQ pain are myriad (especially for adolescents, in whom tubo-ovarian pathology may be confused for appendicitis on an ultrasound). Discordance between a positive ultrasound result for a specific disease and surgical findings is uncommon, but when present, it is usually due to uncommon or late presentations of disease. Studies have revealed that ultrasonography should be the first imaging modality for diagnosing appendicitis in pediatric patients10; however, its sensitivity is lower than that of CT.11 In cases in which the clinical picture is unclear and ultrasonography does not reveal the cause of symptoms, CT or serial abdominal examinations with close monitoring can be used to raise the sensitivity depending on the clinical status of the patient. At our institution, we use a clinical-decision support tool that includes history and examination findings to recommend laboratory testing and imaging (and modality) versus immediate surgical consultation with likely diagnostic laparoscopy.12
Dr Edmunds
Dr Koberlein, what is the false-positive rate of ultrasound imaging for the diagnosis of appendicitis?
Dr Koberlein
Although imaging can have an integral role in the assessment of appendicitis, clinical scoring systems, pathways, and laboratory data are often used to direct when and to what extent imaging is used, as they are at our institution.12 Various published reports have revealed the usefulness and reliability of ultrasonography in the assessment of appendicitis in correlation with examination and laboratory findings, with high specificity and sensitivity seen throughout most published reports.
Despite its general acceptance, there will occasionally be some false-positive examination results when using ultrasonography to assess for appendicitis. Some recent reports have revealed false-positive rates of 11% to 14% while maintaining high sensitivity and specificity.13,14 Possible reasons for higher false-positive rates include incomplete visualization of the appendix, using a strict diameter criteria when deciding if a study result is positive, and mislabeling an adjacent structure as the appendix. Some investigators concluded that demonstrating secondary signs of appendicitis (ie, local inflammation) also should be required to make the diagnosis of acute appendicitis.14
Dr Edmunds
How easy is it to diagnose ovarian torsion in prepubertal children with ultrasound imaging?
Dr Koberlein
Ultrasonography is the most common method of assessing for torsion, and its usefulness in determining torsion has been proven.15 A heterogeneous, asymmetrically enlarged ovary is the most common finding in ovarian torsion, with volumes typically being 12 times that of the normal side.15 Comparison imaging of the contralateral ovary was not done on this patient before she was taken to the operating room. Note that the presence or absence of color flow to the ovaries is not a reliable predictor of torsion.
Conclusions
This patient’s case highlights the importance of a diligent and thorough workup for a patient whose presentation is atypical. The patient’s lack of diarrhea and bradycardia was not typical for simple gastroenteritis; therefore, pursuing more advanced evaluation and imaging was warranted. In young children with vomiting and listlessness, consider ultrasound imaging of the abdomen; although, with strict diagnostic criteria for appendicitis or torsion, a recognition that this imaging modality is not 100% specific is needed. Ovarian torsion should always be on the differential in prepubertal children with vomiting despite its rarity. Frequently, these children will present with a more advanced disease process; therefore, maintaining a high index of suspicion will allow for the determination of an accurate diagnosis and appropriate treatment course.
Footnotes
- Accepted July 12, 2018.
- Address correspondence to Katherine Edmunds, MD, Division of Emergency Medicine, Cincinnati Children’s Hospital, 3333 Burnet Ave, MLC 2008, Cincinnati, OH 45229. E-mail: katherine.edmunds{at}cchmc.org
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.
References
- Copyright © 2019 by the American Academy of Pediatrics
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.