Sage Journals: Discover world-class research

Abstract

Background:

Acute pyelonephritis is a bacterial infection of the upper urinary tract. Patients can be admitted to a variety of wards for treatment. However, at the Prince of Wales Hospital in Hong Kong, they are managed initially in the emergency medicine ward. The aim of the study is to identify the risk factors that are associated with a prolonged hospital length of stay.

Methods:

This was a retrospective cohort study conducted in Prince of Wales Hospital. The study recruited patients who were admitted to the emergency medicine ward between 1 January 2014 and 31 December 2017. These patients presented with clinical features of pyelonephritis, received antibiotic treatment and had a discharge diagnosis of pyelonephritis. The length of stay was measured and any length of stay over 72 h was considered to be prolonged.

Results:

There were 271 patients admitted to the emergency medicine ward, and 118 (44%) had a prolonged hospital length of stay. Univariate and multivariate analyses showed that the only statistically significant predictor of prolonged length of stay was a raised C-reactive protein (odds ratio 1.01; 95% confidence 1.01–1.02; p < 0.0001). Out of 271 patients, 261 received antibiotics in the emergency department. All 10 patients (8.5%) who did not receive antibiotics in emergency department had a prolonged length of stay (p = 0.0002).

Conclusion:

In this series of acute pyelonephritis treated in the emergency medicine ward, raised C-reactive protein levels were predictive for prolonged length of stay. Patients who did not receive antibiotics in the emergency department prior to emergency medicine ward admission had prolonged length of stay.

Keywords

Pyelonephritis sepsis prolonged length of stay infection

Introduction

Acute pyelonephritis is a severe infection involving the upper urinary tract. Clinical presentations of acute pyelonephritis can vary from fever, mild flank discomfort or costovertebral tenderness and urinary tract symptoms, to sepsis or death.^1–4 There are often associated urinary tract symptoms such as dysuria, frequency of micturition, suprapubic pain or urgency. The presence of pyuria or a positive urine culture is traditionally required.⁵ Patients with acute uncomplicated pyelonephritis can be managed in the emergency medicine ward (EMW) within the emergency department (ED). However, some patients can develop complications during their hospital stay, resulting in prolonged hospitalisation. The prognosis of uncomplicated acute pyelonephritis is good,⁶ but complicated pyelonephritis, such as patients with structural or functional abnormalities of the urinary tract, can result in 6%–10% mortality.⁵ The urinary tract is a common source of infection in patients with sepsis and septic shock; these can be life-threatening.^7–9

Several studies have identified the risk factors associated with poor prognosis in patients with complicated acute pyelonephritis. Those risk factors include being elderly (age > 65 years), male sex, bedbound patients, diabetes mellitus, immunosuppressed patients, healthcare-associated infection, obstructive uropathy, reduced platelet count, reduced serum albumin level, increased C-reactive protein (CRP) level, bacteraemia and first episode of acute complicated pyelonephritis.^5,10–13 Another study suggested that the clinical features of acute pyelonephritis did not differ between diabetic and non-diabetic patients, and there was no difference in hospital stay between both groups.¹⁴ Increases in mean platelet volume (MPV) have been suggested to be a predictor for the diagnosis of acute urinary tract infection in children,¹⁵ but other studies have shown no significant association.¹⁶

Uncomplicated acute pyelonephritis patients can be managed in the EMW. An Iranian study suggests that the average EMW length of stay (LoS) is 2.68 days,¹⁷ and if the LoS is longer than 3 days, this could be considered a prolonged stay in the EMW.

We were interested in investigating the predictive factors for a prolonged stay in the EMW for Hong Kong patients presenting with acute pyelonephritis. The aim of this study was to identify any risk factors associated with a prolonged stay in patients with acute pyelonephritis admitted to the EMW.

Methods

This was a single-centre retrospective cohort study which was conducted in Prince of Wales Hospital (PWH), a tertiary university teaching hospital of the Chinese University of Hong Kong. A prolonged hospital LoS was defined as a hospital stay of more than 72 h since the ED registration time.

Ethics

Clinical data were anonymised to protect patient privacy. The requirement for individual patient consent was waived by the Clinical Research Ethics Committee due to the retrospective and anonymous nature of the study. The study was approved by the Joint Chinese University of Hong Kong New Territories East Cluster Clinical Research Ethics Committee. This study was compliant with the Declaration of Helsinki.

Study participants

Patients presenting with acute pyelonephritis who were admitted to the EMW of PWH during the study as described below:

Inclusion criteria

Patients aged ⩾18 years.

EMW admission between 1 January 2014 and 31 December 2017 inclusive.

Received intravenous or oral antibiotic treatment.

The confirmation of discharge diagnosis of pyelonephritis (according to the discharge summary).

Exclusion criteria

Patients aged ⩽18 years.

Patients who have undergone a kidney transplant.

Pregnant women.

Patients who underwent urological surgery within 2 weeks prior to presentation.

Patients presenting with obstructive uropathy, for example, calculus in the urinary tract resulting in hydronephrosis.

Discharge with Acknowledgement of Medical Advice (DAMA) prior to ward admission.

Non-resident of Hong Kong.

Retrieval of data

The clinical data were retrieved from the medical records in the Hospital Authority Clinical Management System (CMS). The clinical data retrieved in this study were patient characteristics, clinical features of pyelonephritis, comorbidities, initial vital signs at triage, blood tests, any radiological imaging, urinalysis, mid-stream urine culture, blood culture, initial antibiotic treatment and length as well as type (intravenous/ oral) of antibiotic treatment which the patients received.

Definitions

The presence of a tympanic temperature of ⩾38°C was defined as pyrexia. The time from patient registration on arrival at the ED of PWH until the patient was discharged from the hospital was defined as the LoS. The expected LoS was less than 72 h (4320 min) from the time of registration of the patient until the time of discharge. Patients who stayed longer than this LoS, that is, ⩾72 h, were defined as having a prolonged hospitalisation. Place of residency was defined as either home or elderly care home. Pre-admission activities of daily living level were described using the modified Rankin Scale (mRS).¹⁸

The definitions of normal blood tests parameters were as follows: white cell count: 4.0–9.7 (×10⁹/L); platelet count: 150–384 (×10⁹/L); MPV: 7.3–10.5 fL; C-CRP: <9.9 mg/l; serum albumin: 33–48 g/L.

Outcome measure

The primary outcome measure was the number of hours of hospital admission. The presence or absence of a prolonged hospital stay was identified. The following predictor variables were used in the analysis: sex; age (>65 years old); activities of daily living; place of residency; diabetes mellitus; white cell count (WCC); CRP; platelet count; MPV and serum albumin.

Data analysis

Continuous variables were dichotomised for data analysis, with univariate analysis for each predictor variable. Continuous data were compared using the Mann–Whitney test as data were not normally distributed whereas categorical data were compared using chi-square tests. All continuous variables are described as median and interquartile range (IQR). A p value of <0.05 was considered statistically significant. Variables with a p value of <0.15 on univariate analysis were included in multivariate logistic regression analysis.

Statistical analyses were performed using MedCalc for Windows, version 19.1 (MedCalc Software, Ostend, Belgium).

Results

Two-hundred and seventy-one patients were included in this study. They all presented to the ED with clinical features of pyelonephritis and were admitted to the EMW in PWH. During this period, 153 patients (56.5%) were admitted to the EMW for less than 72 h, whereas 118 patients (43.5%) were admitted for more than 72 h.

The median age of patients was 45 years old (IQR: 33–56), and 30 patients (11.1%) were ⩾65 years old. The majority of patients were female (256, 94.5%). Most patients were independently mobile (mRS 0, 253 patients, 93.4%). The recorded mRS was 1 for 16 patients (5.9%) and 3 for 2 patients (0.7%). The median hospital LoS was 69 h (IQR: 50–96). Thirty patients (11.1%) developed complications (including hydronephrosis or obstruction) while staying in EMW and were admitted to other specialties. Fourteen patients (5.2%) were admitted to urology, nine patients (3.3%) were admitted to internal medicine, five patients (1.9%) were admitted to general surgery, and two patients (0.7%) were admitted to the intensive care unit.

Table 1 shows the comparison of variables between <72 h LoS and ⩾72 h LoS. Table 2 shows the odds ratios of predictors for prolonged hospitalisation. Results showed that increased CRP was significantly associated with a prolonged LoS. Multiple logistic regression analysis confirmed that the only statistically significant predictor of prolonged LoS was a raised CRP (Table 2, p < 0.0001).

Table 1.

Comparison between with and without prolong hospitalisation in 271 patients with acute pyelonephritis.

	Hospital ⩽ 72 h (n = 153)		Hospital > 72 h (n = 118)		p value
Basic characteristics
Age	44	[32–55]	46	[33–60]	0.38
Age ⩾ 65 years old	13	(8.5%)	17	(14.4%)	0.12
Sex (female)	146	(95.4%)	110	(93.2%)	0.43
DM	10	(6.5%)	15	(12.7%)	0.08
Clinical presentation
Triage temperature (°C)	38.0	[37.6–39.0]	38.4	[37.6–39.3]	0.11
Fever	119	(77.8%)	92	(78.0%)	0.97
Loin pain tenderness	146	(95.4%)	106	(89.8%)	0.07
Urinary tract symptoms
Yes	115	(75.2%)	83	(70.3%)	0.38
No	38	(24.8%)	35	(29.7%)
Dysuria
Yes	103	(67.3%)	70	(59.3%)	0.18
No	50	(32.7%)	48	(40.7%)
Urgency
Yes	18	(11.8%)	8	(6.8%)	0.17
No	135	(88.2%)	110	(93.2%)
Frequency
Yes	64	(41.8%)	42	(35.6%)	0.30
No	89	(58.2%)	76	(64.4%)
Haematuria
Yes	32	(20.9%)	17	(14.4%)	0.17
No	121	(79.1%)	101	(85.6%)
WBC positive urinalysis
Yes	15	(9.8%)	31	(26.3%)	0.76
No	108	(70.6%)	95	(80.5%)
Nitrite positive urinalysis
Yes	12	(7.8%)	23	(19.5%)	0.17
No	52	(34.0%)	48	(40.7%)
Mid-stream urine culture
Positive	44	(28.8%)	40	(33.9%)	0.18
No test performed	52	(34.0%)	27	(22.9%)
No bacterial growth	46	(30.1%)	38	(32.2%)
Laboratory characteristics
Haemoglobin (g/dL)	12.7	[11.8–13.4]	12.7	[11.6–13.7]	0.84
WBC count (×10⁹/L)	11.6	[9.3–14.0]	13.5	[10.8–16.3]	0.0003
Platelet count (×10⁹/L)	220	[181–266]	196	[153–243]	0.0006
Mean platelet volume (fL)	8.5	[8.1–9.1]	8.6	[8.0–9.1]	0.83
Urea (mmol/L)	3.8	[3.1–4.9]	4.1	[3.3–5.5]	0.08
Creatinine (µmol/L)	66	[59–73]	71	[63–90]	<0.0001
C-reactive protein (mg/L)	44.0	[18.8–95.5]	95.1	[52.7–168.6]	<0.0001
Serum albumin (g/L)	40	[36–43]	39	[35–41]	0.07
Neutrophil level	9.4	[6.9–11.6]	11.2	[8.8–14.4]	<0.0001
Neutrophil percentage (%)	79	[73–86]	84	[78–89]
Antibiotic management
Antibiotic administration in ED
Yes	153	(100%)	108	(91.5%)	0.0002
No	0	(0%)	10	(8.5%)
Types of antibiotics
Ceftriaxone	144	(94.1%)	92	(85.2%)	0.05
Co-amoxiclav	5	(3.3%)	10	(9.3%)
Others	4	(2.6%)	6	(5.5%)
Initial route of antibiotic administration
Intravenous	151	(98.7%)	107	(99.1%)	0.77
Oral	2	(1.3%)	1	(0.9%)
Change of antibiotics
Yes	30	(19.6%)	46	(39.0%)	0.0001
No	123	(80.4%)	61	(51.7%)

WBC: white blood cell; ED: emergency department; DM: Diabetes Mellitus.

Table 2.

Logistic regression model of factors for predicting prolonged hospitalisation.

Variable	Enter
	Odds ratio	95% CI	p value
Age	1.00	0.98–1.03	0.77
Sex (female)	0.37	0.074–1.83	0.22
Diabetes mellitus	2.19	0.66–7.27	0.20
Temperature at triage (°C)	1.20	0.90–1.60	0.22
Loin pain/ tenderness	0.39	0.081–1.90	0.25
Urinary tract symptoms	0.83	0.38–1.83	0.65
C-reactive protein (mg/L)	1.01	1.01–1.02	<0.0001
Neutrophil level	1.03	0.96–1.11	0.43

CI: confidence interval.

Out of 271 patients, 261 received antibiotics in the ED. Ten patients (8.5%) did not receive antibiotic treatment prior to EMW admission; this was associated with prolonged LoS (Table 1, p = 0.0002). In the 261 patients who received antibiotic treatment, 236 patients (87.1%) received ceftriaxone, 15 patients (5.5%) received co-amoxiclav and 10 patients (3.7%) received other antibiotics. Intravenous antibiotics were given to 258 patients (95.2%) in ED and 3 patients (1.1%) received oral antibiotics. Seventy-seven patients (29.1%) had a change of antibiotics from their initial regimen and 186 patients (70.2%) received the same antibiotic regimen throughout their stay.

Results of mid-stream urine sensitivities were obtained in 234 patients (86.3%), and no test was performed in 37 patients (13.7%). Escherichia coli species were detected in the majority of cases (74 patients, 27.3%), and one sample revealed Extended Spectrum Beta Lactamase (ESBL) E.coli. Three patients had Enteroccocus species (1.1%) and four patients (1.5%) showed mixed growth or contamination.

Discussion

We investigated different factors that might lead to a prolonged LoS in patients presenting with acute pyelonephritis to our EMW. Our study found that a raised CRP level was associated with a prolonged hospital LoS for patients with acute pyelonephritis, and this was statistically significant after multiple logistic regression. The CRP assay is available in our hospital at all times and has a turnaround time of 2–4 h; it costs around HK$135 per assay (personal communication, Dr Michael Chan, Department of Chemical Pathology).

Several studies have established that raised CRP levels are associated with increased mortality in pyelonephritis.^5,10–13 High CRP levels have been shown to be an independent predictor of the need for urinary drainage, as well as an objective and useful parameter for the selection of stenting in patients with renal colic.¹⁸ Angulos et al. investigated CRP levels as a predictor of the need for urgent urinary diversion in patients with renal colic and urolithiasis. They concluded that CRP levels provided an objective and useful parameter for urinary stent placement.¹⁹ Raised CRP levels were related to the development of complications, which is consistent with CRP levels being an associated predictive factor for prolonged LoS.

Our study also suggests that patients who received antibiotics in the ED prior to EMW admission had a shortened LoS in comparison to those who did not receive antibiotics in ED. Several studies have established that empirical intravenous antibiotic treatment should be initiated early in urosepsis²⁰ and more generally.²¹ Third-generation cephalosporins are reported to be the initial empirical antibiotic treatment of choice in elderly patients with urosepsis,²² but care should be taken to ensure that patients are not allergic to these agents or to penicillin prior to administration.²³ Dreger et al²⁴ have established that each additional hour of delay in initiating antibiotic treatment in urosepsis could lower the survival rate by 7.6%. Our study shows that delays in antibiotic treatment can prolong hospital LoS which is supported by other studies of the importance of early antibiotic administration.^20,22,24

The strengths of this study include the sample size of 271 patients and the consistency of data collection over a 4-year period. However, this study is a single-centre retrospective study, and all retrospective studies are prone to bias. In particular, they can only demonstrate associations and not causality. There could be diagnostic bias, selection bias and recall bias in this study as a result. In particular, some patients with acute pyelonephritis may have been admitted directly to general medicine or urology wards rather than the EMW, so some cases may have been missed by our methodology. Nevertheless, this study has demonstrated the possible utility of raised CRP levels and the usefulness of early intervention by initiating antibiotic treatment in the ED, with a potentially positive effect on reduction of hospital LoS. Future studies could investigate the role of CRP further and explore the utility of other potential biomarkers for renal disease.²⁵

Summary

Raised CRP levels could be a potential predictor for prolonged LoS for patients with acute pyelonephritis. Patients with acute uncomplicated pyelonephritis can be managed in the EMW, and early initiation of intravenous antibiotics in ED prior to admission may reduce the hospital LoS.

Footnotes

The authors would like to thank all the patients and colleagues in the Emergency Department of the Prince of Wales Hospital,Hong Kong. They also thank the staff of the Accident and Emergency Medicine Academic Unit at the Chinese University of Hong Kong.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research,authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research,authorship and/or publication of this article.

Availability of data and materials

The anonymous data which form the basis for this study are available from the authors on reasonable request.

Ethical approval and Human rights statement

Clinical data were anonymised to protect patient privacy. The requirement for individual patient consent was waived by the Clinical Research Ethics Committee due to the retrospective and anonymous nature of the study. The study was approved by the Chinese University of Hong Kong – Prince of Wales Hospital Clinical Research Ethics Committee. This study was compliant with the Declaration of Helsinki.

Informed consent

The requirement for individual patient consent was waived by the Clinical Research Ethics Committee due to the retrospective and anonymous nature of the study.

ORCID iDs

Howard Tat Chun Chan

Alex Kwok Keung Law

Colin A Graham

References

Czaja

Scholes

Hooton

, et al. Population-based epidemiologic analysis of acute pyelonephritis. Clin Infect Dis 2007: 45273–45280.

Scholes

Hooton

Roberts

, et al. Risk factors associated with acute pyelonephritis in healthy woman. Ann Intern Med 2005: 14220–14227.

Van Nieuwkoop

van’t Wout

Spelt

, et al. Prospective cohort study of acute pyelonephritis in adults: safety of triage towards home based oral antimicrobial treatment. J Infect 2010; 60(2): 114–121.

Horcajada

Shaw

Padilla

, et al. Healthcare-associated, community acquired bacteraemic urinary tract infections in hospitalized patients: a prospective multicentre cohort study in the era of antimicrobial resistance. Clin Microbol Infect 2013; 19: 962–968.

Ruiz-Mesa

Marquez-Gomez

Sena

, et al. Factors associated with severe sepsis or septic shock in complicated pyelonephritis. Medicine 2017; 96: 43.

Velasco

Martinez

Moreno-Martinez

, et al. Blood cultures for women with uncomplicated acute pyelonephritis: are they necessary? Clin Infect Dis 2003; 37: 1127–1130.

Blanco

Muriel-Bombin

Sagredo

, et al. Incidence, organ dysfunction and mortality in severe sepsis: a Spanish multicentre study. Crit Care 2008; 12(6): R158.

Seymour

Rea

Kahn

, et al. Severe sepsis in pre-hospital emergency care: analysis of incidence, care, and outcome. Am J Respir Crit Care Med 2012; 186: 1264–1271.

Gray

Ward

Lees

, et al. The epidemiology of adults with severe sepsis and septic shock in Scottish emergency departments. Emerg Med J 2013; 30(5): 397–401.

10.

Lee

Cho

JH.

Risk factors of septic shock in bacteremic acute pyelonephritis patients admitted to an ER. J Infect Chemother 2012; 18(1): 130–133.

11.

Yamamoto

Fujita

Nakazawa

, et al. Clinical characteristics and risk factors for septic shock in patients receiving emergency drainage for acute pyelonephritis with upper urinary tract calculi. BMC Urol 2012; 12: 4.

12.

Tambo

Okegawa

Shishido

, et al. Predictors of septic shock in obstructive acute pyelonephritis. World J Urol 2014; 32(3): 803–811.

13.

Hsu

Fang

Chou

, et al. The clinical impact of bacteraemia in complicated acute pyelonephritis. Am J Med Sci 2006; 332: 17580.

14.

Trivedi

Phatak

Trivedi

Retrospective comparison of clinical characteristics and in-hospital outcomes among diabetic and non-diabetic adults with acute pyelonephritis. J Clin Diagn Res 2016; 10(10): OC26–OC29.

15.

Tekin

Konca

Gulyuz

, et al. Is the mean platelet volume a predictive marker for the diagnosis of acute pyelonephritis in children. Clin Exp Nephrol 2015; 19(4): 688–693.

16.

Lee

Shin

Park

, et al. Mean platelet volume in young children with urinary tract infection. Sci Rep 2015; 5: 18072.

17.

Salehi

Nayeri

Rashidian

, et al. Investigating the effect of clinical governess approach on patients’ length of stay in emergency department: an action research study. Acta Med Iran 2014; 52(2): 137–145.

18.

Broderick

Adeoye

Elm

Evaluation of the Modified Rankin Scale and its use in future stroke trials. Stroke 2017; 48(7): 2007–2012.

19.

Angulo

Gaspar

Rodriguez

, et al. The value of C-reactive protein determination in patients with renal colic to decide urgent urinary diversion. Urology 2010; 76(2): 301–306.

20.

Wagenlehner

FME

Pilatz

Weidner

, et al. Urosepsis: overview of the diagnostic and treatment challenges. Microbiol Spectr 2015; 3(5): 0003.

21.

Chang

C-P

S-R

Lin

C-N.

No better outcome of using antibiotic broadly in hospitalized adult patients with leukocytosis in emergency department. Hong Kong J Emer Med 2020; 27(2): 87–91.

22.

Heppner

Yapan

Wiedemann

Urosepsis in geriatric patients. Aktuelle Urol 2016; 47(1): 54–59.

23.

Aiyaka

Techakehakij

The prevalence of cross-reactivity of cephalosporin in penicillin-allergic patients: a cross-sectional study in Thailand. Hong Kong J Emer Med 2019; 26(3): 151–155.

24.

Dreger

Degener

Ahmad-Nejad

, et al. Urosepsis: etiology, diagnosis and treatment. Dtsch Arztebl Int 2015; 112(49): 837–847.

25.

Işıkkent

Yılmaz

Özturan

İU

, et al. Utility of neutrophil gelatinase-associated lipocalin in the management of acute kidney injury: a prospective, observational study. Hong Kong J Emer Med 2020; 27(1): 8–14.

Predictive factors for prolonged hospitalisation in acute pyelonephritis patients admitted to the emergency medicine ward