The present study investigated the incidence of hyperammonemia in urinary tract infections and explored the utility of urinary obstruction relief and antimicrobial administration to improve hyperammonemia.
This was an observational study. Subjects were patients who were diagnosed with urinary tract infection and hospitalized between June 2008 and June 2009. We measured plasma ammonia levels on admission in patients who were clinically diagnosed with urinary tract infection and hospitalized. We assessed each patient's level of consciousness on admission using the Glasgow Coma Scale (GCS) and performed urine and blood cultures. We also assessed hearing prior to hospitalization using the Eastern Cooperative Oncology Group performance status (ECOG-PS). In cases with high ammonia levels on admission, plasma ammonia and GCS were measured 24 hours and 5–7 days later.
Sixty-seven candidates were enrolled; of these, 60 cases (89.6%) with bacterial cell counts ≥104 CFU/mL were studied. Five cases (8.3%) presented with high plasma ammonia levels. Cases with hyperammonemia were significantly more likely to present with low GCS scores and urinary retention rate. All five cases received antimicrobial therapy with an indwelling bladder catheter to relieve urinary retention. The case 5 patient died shortly after admission due to complicated aspiration pneumonia; in the remaining cases, plasma ammonia levels were rapidly normalized and the level of consciousness improved.
Citation: Kenzaka T, Kato K, Kitao A, Kosami K, Minami K, Yahata S, et al. (2015) Hyperammonemia in Urinary Tract Infections. PLoS ONE 10(8): e0136220. https://doi.org/10.1371/journal.pone.0136220
Editor: Jeff M. Sands, Emory University, UNITED STATES
Received: April 17, 2015; Accepted: July 31, 2015; Published: August 20, 2015
Copyright: © 2015 Kenzaka et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are presented in the paper. Also, data are available from the Department of General Medicine, Toyooka Public Hospital Institutional Data Access upon request. The anonymized patient data cannot be made publicly available due to ethical restrictions implemented by Toyooka Public Hospital Institutional Data Access. Requests for the data may be sent to Mr. Yasuyuki Hoshiba; Department of general affairs, Toyooka Public Hospital, Toyooka, Japan. E-mail: email@example.com TEL:+81-796-22-6111, FAX:+81-796-22-0170.
Funding: The authors received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Hyperammonemia is caused by hepatic disorder/failure, gastrointestinal hemorrhage, portosystemic shunt, and vesicorectal fistula, drugs such as valproate and barbiturate, and shock . Obstructive urinary tract infection with urease-producing bacteria also causes hyperammonemia , a rare pathologic condition whose incidence is unknown.
Treatment of hepatogenic hyperammonemia includes restriction of protein intake and the use of nonabsorbable disaccharides such as lactulose and lactitol to reduce intestinal ammonia. In contrast, there are no guidelines for standard therapy for hyperammonemia due to obstructive urinary tract infection , but relief of urinary retention improves disorders of consciousness and hyperammonemia .
The present study investigated the incidence of hyperammonemia in urinary tract infections and explored the utility of urinary obstruction relief and antimicrobial administration alone to improve hyperammonemia.
This was an observational study conducted at the Department of General Medicine of Toyooka Public Hospital, Hyogo, Japan. The local ethics committee of Toyooka Public Hospital approved the study and waived the requirement for written informed consent only for participants 18 years of age or older. The ethical committee required minors (and/or their parents/guardians) to provide written informed consent. All data are available upon request; relevant data are presented here.
Study Setting and Population
Subjects were patients who were diagnosed with urinary tract infection and hospitalized between June 2008 and June 2009. The diagnosis of urinary tract infection was based on the appendix of the fourth edition of the “Criteria for evaluation of clinical efficacy of antimicrobial agents on urinary tract infection” by the UTI Subcommittee of the Clinical Evaluation Committee, Japan Society of Chemotherapy . The cases that met the following (1) to (3) conditions were enrolled as candidates; cases that met condition (4) were comprised the study subjects: (1) age ≥16 years; (2) fever ≥37.5°C or low back, flank, or kidney pain; (3) pyuria ≥5 WBCs/high power field in the microscopic urine sediment; and (4) bacterial cell counts ≥104 colony-forming units (CFU)/mL.
Exclusion criteria included diagnoses of urethritis, prostatitis, or epididymitis; presence of an indwelling bladder balloon; complications of the following diseases that cause hyperammonemia: hepatic disorder/failure, gastrointestinal hemorrhage, portosystemic shunt, urea cycle disorder, or vesicorectal fistula; and treatment with valproate, barbiturate, narcotic, diuretic, or cancer chemotherapy.
We measured plasma ammonia levels on admission in patients who were clinically diagnosed with urinary tract infection and hospitalized with conditions (1) to (3) at the time of medical examination. We assessed each patient’s level of consciousness on admission using the Glasgow Coma Scale (GCS) and performed urine and blood cultures. We also assessed hearing prior to hospitalization using the Eastern Cooperative Oncology Group performance status (ECOG-PS) (Table 1). In cases with high ammonia levels on admission, plasma ammonia and GCS were measured 24 hours and 5–7 days later.
All patients received standard antimicrobial therapy. Empirically, patients with no or low risk factors for drug-resistant pathogens were treated with a non-antipseudomonal beta-lactam antibiotic such as ceftriaxone or cefazolin. Patients with high risk factors for drug-resistant pathogens were treated with antipseudomonal beta-lactam antibiotics such as piperacillin-tazobactam, cefepime, or meropenem, plus vancomycin for Enterococcus species in some cases. Based on culture results, patients received subsequent definitive therapy.
On admission, venous blood samples were collected in heparin collection tubes and immediately used for measurement. Ammonia levels were determined using a sensitive immunolabeling assay reagent, Serotec Ammonia-L (Serotec, Hokkaido, Japan). The normal range for plasma ammonia is 12–66 μg/dL.
Baseline characteristics were compared in patients with and without hyperammonemia. Statistical significance was determined by t test or Chi-square test. All statistical analyses were performed using IBM SPSS for Windows version 22.0 (IBM Inc., New York, NY). The significance level was set at P < 0.05 for all tests.
Sixty-seven candidates were enrolled; of these, 60 cases (89.6%) with bacterial cell counts? 104 CFU/mL were studied. Table 2 shows demographic characteristics of the subjects. Subjects included 21 men and 39 women. The mean age was 76.3 ± 14.9 years (range 19 to 99 years old). ECOG-PS classed 14 cases (23.3%) as grade 0, 14 cases (23.3%) as grade 1, 11 cases (18.3%) as grade 2, eight cases (13.3%) as grade 3, and 13 cases (21.7%) as grade 4. Eighteen cases (30.0%) had a disorder of consciousness with GCS scores <14. Eight cases had urinary retention (13.3%). Twenty-three cases (38.3%) were blood culture-positive for bacteria. Five cases (8.3%) presented with high plasma ammonia levels. Cases with hyperammonemia were significantly more likely to present with low GCS scores and urinary retention rate.
Results of urine cultures are shown in Table 3. Escherichia coli was detected in 30 cases (50.0%), followed by Klebsiella pneumonia in five cases (8.3%), Proteus mirabilis in five cases (8.3%), and Streptococcus agalactiae in four cases (6.7%).
The clinical characteristics and changes in plasma ammonia levels in cases with hyperammonemia are shown in Tables 4 and 5. Their ECOG-PS grades were 2–4 and they were not necessarily bedridden, but all had urinary retention. All five cases received antimicrobial therapy and an indwelling bladder catheter to relieve urinary retention. The case 5 patient died shortly after admission due to complicated aspiration pneumonia; in the remaining cases, plasma ammonia levels were rapidly normalized and the level of consciousness improved (cases 1–4).
In this study, the incidence of hyperammonemia and disorders of consciousness caused by hyperammonemia was 8.3% in 60 targeted cases with urinary tract infections. There has been only prior report of a case of hyperammonemia caused by obstructive urinary tract infection and it is thus considered a rare pathologic condition , although we observed a relatively high incidence of this condition. Our study demonstrated that hyperammonemia caused by obstructive urinary tract infection is not a rare pathologic condition in elderly with poor performance status, of which the mean age was 76.3 ± 14.9 years; patients with ECOG-PS grade ≥2 accounted for 55.0% of these cases. Cases with hyperammonemia were significantly more likely to present with low GCS scores and urinary retention. When a case with a disorder of consciousness accompanies hyperammonemia, an obstructive urinary tract infection should be included in differential diagnosis.
Cases 1–4 with hyperammonemia accompanying obstructive urinary tract infections received antimicrobial therapy and an indwelling bladder catheter to relieve urinary retention, which resulted in rapid normalization of plasma ammonia levels and improved consciousness. The onset of hyperammonemia caused by urinary tract infections may occur as a result of elevated intravesical pressure due to dysuria caused by neurogenic bladder or prostatic hypertrophy. Then, urinary ammonia is absorbed by the vesical venous plexus and transferred to the systemic circulation through the inferior vena cava, instead of being transported via the liver through the internal iliac vein . In addition, urease-producing bacteria in the bladder likely promote hyperammonemia [5,6].
When the urinary tract is infected with urease-producing bacteria, urease hydrolyzes urea in the urine to ammonium ions (NH4+), which elevate urinary pH. On the other hand, as ammonium ions become lipophilic ammonia (NH3) in the alkaline urine, it is easily transferred to the vesical venous plexus . Hyperammonemia may develop due to urinary retention without infection by urease-producing bacteria . P. mirabilis, K. pneumonia, and Pseudomonas aeruginosa are common urease-producing bacteria associated with urinary tract infections [8,9]. However, all bacteria causing hyperammonemia in this study were species that rarely produce urease  and urease activity was not detected in any of the cultured bacteria. The results of this study suggest that even without involvement of urease-producing bacteria, the fact that intravesical pressure is elevated and urinary ammonia is absorbed by the vesical venous plexus may cause hyperammonemia and accompanying disorders of consciousness. Therefore, along with antimicrobial administration, relief of obstruction is an effective treatment to reduce intravesical pressure and prevent absorption of urinary ammonia by the vesical venous plexus.
We targeted only hospitalized patients and observed the incidence of hyperammonemia in an elderly group with relatively poor performance status. Cases who had received antimicrobial therapy and exhibited cell counts <104 CFU/mL at the time of urine testing were excluded; thus, the incidence in the general population with urinary tract infections is likely lower. This study included only five cases with hyperammonemia. To strengthen our conclusions, future studies should include a larger sample and a group of patients with hyperammonemia but who do not have a bladder catheter.
The occurrence of hyperammonemia in urinary tract infections is not rare. The elevation of intravesical pressure due to urinary retention, followed by the absorption of urinary ammonia by the vesical venous plexus leads to hyperammonemia. Therefore, along with antimicrobial administration, relief of obstruction is important for the treatment of hyperammonemia being caused by this mechanism.
Conceived and designed the experiments: TK MO. Performed the experiments: TK K. Kato AK K. Kosami KM SY MF. Analyzed the data: TK MF. Contributed reagents/materials/analysis tools: TK K. Kato AK K. Kosami KM SY MF. Wrote the paper: TK SY MF.
- 1. Hawkes ND, Thomas GA, Jurewicz A, Williams OM, Hillier CE, McQueen IN, et al. Non-hepatic hyperammonaemia: an important, potentially reversible cause of encephalopathy. Postgrad Med J. 2001;77:717–722. pmid:11677282
- 2. De Jonghe B, Janier V, Abderrahim N, Hillion D, Lacherade JC, Outin H. Urinary tract infection and coma. Lancet. 2002;360:996. pmid:12383670
- 3. Kenzaka T, Kumabe A, Urushibara Y, Minami K, Ishida T. Hyperammonemia in a patient with obstructive urinary tract infection due to Corynebacterium urealyticum. Cent Eur J Med. 2013;8:597–599.
- 4. Criteria for evaluation of clinical efficacy of antimicrobial agents on urinary tract infection. UTI Subcommittee of the Clinical Evaluation Committee, Japan Society of Chemotherapy. Jpn J Antibiot. 2000;53:453–477. pmid:10955240
- 5. Kaveggia FF, Thompson JS, Schafer EC, Fischer JL, Taylor RJ. Hyperammonemic encephalopathy in urinary diversion with urea-splitting urinary tract infection. Arch Intern Med. 1990;150:2389–2392. pmid:2241451
- 6. Samtoy B, DeBeukelaer MM. Ammonia encephalopathy secondary to urinary tract infection with Proteus mirabilis. Pediatrics. 1980;65:294–297. pmid:6444460
- 7. Oliver RM, Talbot S, Raman GV. Hyperammonaemic coma in ureterosigmoid urinary diversion. Postgrad Med J. 1989;65:502–504. pmid:2602245
- 8. Albersen M, Joniau S, Van Poppel H, Cuyle PJ, Knockaert DC, Meersseman W. Urea-splitting urinary tract infection contributing to hyperammonemic encephalopathy. Nat Clin Pract Urol. 2007;4:455–458. pmid:17673917
- 9. Arai Y, Takeuchi H, Tomoyoshi T, Tatewaki K. Urease activity of bacteria in urine. Hinyokika Kiyo. 1989;35:277–281. pmid:2500012