U r i n a r y Tr a c t I n f e c t i o n s i n S p e c i a l Populations Diabetes, Renal Transplant, HIV Infection, and Spinal Cord Injury Lindsay E. Nicolle,
MD, FRCPC
KEYWORDS Urinary infection Cystitis Pyelonephritis Asymptomatic bacteriuria Diabetes Renal transplant HIV infection Spinal cord injury KEY POINTS Patients with diabetes are more likely to present with complications of urinary infection, such as abscesses and emphysematous cystitis or pyelonephritis. Renal transplant patients have a high frequency of urinary infection because of multiple risk factors that may predate transplant, are associated with technical aspects of transplant surgery, or follow transplant. There is limited, if any, increased frequency of urinary tract infection directly attributable to HIV infection. Prevention of urinary tract infections in individuals with spinal cord injuries requires appropriate bladder management to maintain a low-pressure bladder, and avoidance of indwelling devices if possible.
INTRODUCTION
Some populations have unique considerations relevant to urinary tract infection. This article addresses 4 of these groups: patients with diabetes, patients with a renal transplant, patients with HIV infection, and patients with a spinal cord injury. Urinary tract infection occurring in these individuals is considered within the clinical category of complicated urinary infection; that is, infection that occurs in a patient with functional or structural abnormalities of the genitourinary tract. It is always important to distinguish between symptomatic urinary infection and asymptomatic infection, also referred to as bacteriuria, for optimal management of infection.
Health Sciences Centre, Room GG443, 820 Sherbrook Street, Winnipeg, Manitoba R3A 1R9, Canada E-mail address: [email protected] Infect Dis Clin N Am 28 (2014) 91–104 http://dx.doi.org/10.1016/j.idc.2013.09.006 0891-5520/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
id.theclinics.com
92
Nicolle
PATIENTS WITH DIABETES Unique Aspects of Urinary Infection
It is generally accepted that persons with diabetes have an increased frequency of urinary infection,1,2 but there is limited evidence confirming the magnitude of excess risk.3 In addition, the diabetic population is heterogeneous and the risk of urinary infection varies with patient characteristics. Several explanations have been proposed to explain an increased risk for infection, including glucosuria and impaired immune or leukocyte function,2 but experimental studies have not consistently supported any single mechanism. The important diabetes-specific risk factors for urinary infection are usually duration of diabetes or presence of long-term complications, such as neuropathy, rather than current glucose control (Table 1).4 There is limited evidence describing aspects of urinary infection in diabetic men. Of interest, the SGLT2 (serum glucose cotransporter-2) inhibitors, a new class of agents for treatment of diabetes that produce high levels of glucosuria, are associated with only a small increase in symptomatic urinary infection for both men and women.7 Epidemiology
Rates of urinary infection were compared between diabetic women enrolled in the epidemiology of diabetes interventions and complications study (Uro-EDIC) and nondiabetic women in the National Health and Nutrition Examination Survey III. The adjusted prevalence of cystitis in the preceding 12 months was similar (odds ratio [OR] 0.78; 95% confidence interval [CI]: 0.51, 1.22).8 In the Uro-EDIC study, only sexual intercourse was associated with cystitis (OR 8.28; 95% CI 1.45, 158.32), similar to the nondiabetic population. Neither cystitis nor pyelonephritis was associated with duration of diabetes, hemoglobin A1c, retinopathy, neuropathy, nephropathy, vascular complications, or glycemic therapy. A prospective study from the Netherlands also reported that only sexual intercourse was associated with symptomatic infection in women with type 1 diabetes (relative risk [RR] 3.6; P 5 .004), whereas asymptomatic bacteriuria was the only significant association for type 2 diabetes (RR 1.65; 95% CI 1.02, 2.67).5 Another prospective study enrolling women in a US health maintenance organization reported increased symptomatic urinary infection in postmenopausal women with diabetes (OR 2.2; 95% CI 1.6–3.0) for subjects receiving oral diabetes medication or insulin.9 A retrospective record review of patients attending primary care practices in the Netherlands reported recurrent urinary infection was increased for women with diabetes (OR 2.0; 95% CI 1.4–2.9).4 The increased risk was independently associated with type 2 diabetes, diabetes of 5 or more years’ duration, receiving oral or insulin therapy, or retinopathy. Hemoglobin A1c was not a risk factor. Studies that have used administrative databases or retrospective record
Table 1 Variables associated with symptomatic urinary tract infection or asymptomatic bacteriuria in women with Type 2 diabetes Risk Factors for Infection (Ref.) Symptomatic4,5
Asymptomatic6
Not diabetes associated
Age
Age
Diabetes associated
Retinopathy Oral hypoglycemic or insulin therapy Diabetes 5 y
Any long-term complication Heart disease Duration of diabetes Oral hypoglycemic therapy
Urinary Tract Infections in Special Populations
review, however, may overestimate the frequency of urinary infection in diabetic women, because these patients are more likely to seek medical care than nondiabetic persons. Diabetes is a common risk factor associated with more severe presentations of urinary tract infection. A case series of 65 consecutive patients with renal or perinephric abscesses reported 28% of subjects had diabetes mellitus.10 In other case series, 67% of patients presenting with emphysematous cystitis were diabetic,11,12 and 62% presented with emphysematous pyelonephritis.11 Diabetes mellitus has not, however, been identified as a risk factor for complications of severe sepsis or septic shock in patients with urosepsis.13 The prevalence of asymptomatic bacteriuria is increased for diabetic women but not diabetic men, with reported rates of 5% to 25% for women and 3% for men.6 Asymptomatic bacteriuria correlates with duration of diabetes and long-term complications of diabetes, but not with parameters of current metabolic control (see Table 1). Management Diagnosis
The clinical and microbiologic diagnosis of urinary tract infection in diabetic populations is similar to other patients with complicated infection. Patients with neuropathy may have impaired bladder sensation, which obscures some clinical symptoms of lower tract infection. Diabetic women with pyelonephritis are more likely to have bilateral renal involvement and bacteremia. In a group of elderly Greek patients hospitalized with pyelonephritis, bacteremia was identified in 30.7% with and 11% without diabetes.14 The types of infecting organisms are also similar for diabetic and nondiabetic patients.15,16 Escherichia coli is the single most common organism. Strains isolated from women with or without diabetes have similar virulence characteristics.3 Treatment
Diabetic women with a functionally normal genitourinary tract are managed similarly to other women with uncomplicated urinary infection. Cystitis should be treated with short-course therapy and pyelonephritis for 7 to 14 days.17 Complicated infections may require more prolonged antimicrobial therapy.16 Imaging is indicated for severe clinical presentations, failure to respond to therapy, or early symptomatic recurrence following discontinuation of antimicrobial therapy. Emphysematous pyelonephritis is managed with antimicrobial therapy and initial percutaneous drainage; delayed nephrectomy, where necessary, is performed once the patient is stable.11 A perinephric abscess usually requires percutaneous or open drainage. Small renal abscesses of less than 3 cm in diameter are treated conservatively with antimicrobial therapy alone and continued until the abscess has resolved on follow-up imaging.10 This treatment may require prolonged therapy for several weeks to months. Outcomes
Hospitalized elderly Greek diabetic patients with acute pyelonephritis had a longer duration of fever (median 4.5 vs 2.5 days, P
MD, FRCPC
KEYWORDS Urinary infection Cystitis Pyelonephritis Asymptomatic bacteriuria Diabetes Renal transplant HIV infection Spinal cord injury KEY POINTS Patients with diabetes are more likely to present with complications of urinary infection, such as abscesses and emphysematous cystitis or pyelonephritis. Renal transplant patients have a high frequency of urinary infection because of multiple risk factors that may predate transplant, are associated with technical aspects of transplant surgery, or follow transplant. There is limited, if any, increased frequency of urinary tract infection directly attributable to HIV infection. Prevention of urinary tract infections in individuals with spinal cord injuries requires appropriate bladder management to maintain a low-pressure bladder, and avoidance of indwelling devices if possible.
INTRODUCTION
Some populations have unique considerations relevant to urinary tract infection. This article addresses 4 of these groups: patients with diabetes, patients with a renal transplant, patients with HIV infection, and patients with a spinal cord injury. Urinary tract infection occurring in these individuals is considered within the clinical category of complicated urinary infection; that is, infection that occurs in a patient with functional or structural abnormalities of the genitourinary tract. It is always important to distinguish between symptomatic urinary infection and asymptomatic infection, also referred to as bacteriuria, for optimal management of infection.
Health Sciences Centre, Room GG443, 820 Sherbrook Street, Winnipeg, Manitoba R3A 1R9, Canada E-mail address: [email protected] Infect Dis Clin N Am 28 (2014) 91–104 http://dx.doi.org/10.1016/j.idc.2013.09.006 0891-5520/14/$ – see front matter Ó 2014 Elsevier Inc. All rights reserved.
id.theclinics.com
92
Nicolle
PATIENTS WITH DIABETES Unique Aspects of Urinary Infection
It is generally accepted that persons with diabetes have an increased frequency of urinary infection,1,2 but there is limited evidence confirming the magnitude of excess risk.3 In addition, the diabetic population is heterogeneous and the risk of urinary infection varies with patient characteristics. Several explanations have been proposed to explain an increased risk for infection, including glucosuria and impaired immune or leukocyte function,2 but experimental studies have not consistently supported any single mechanism. The important diabetes-specific risk factors for urinary infection are usually duration of diabetes or presence of long-term complications, such as neuropathy, rather than current glucose control (Table 1).4 There is limited evidence describing aspects of urinary infection in diabetic men. Of interest, the SGLT2 (serum glucose cotransporter-2) inhibitors, a new class of agents for treatment of diabetes that produce high levels of glucosuria, are associated with only a small increase in symptomatic urinary infection for both men and women.7 Epidemiology
Rates of urinary infection were compared between diabetic women enrolled in the epidemiology of diabetes interventions and complications study (Uro-EDIC) and nondiabetic women in the National Health and Nutrition Examination Survey III. The adjusted prevalence of cystitis in the preceding 12 months was similar (odds ratio [OR] 0.78; 95% confidence interval [CI]: 0.51, 1.22).8 In the Uro-EDIC study, only sexual intercourse was associated with cystitis (OR 8.28; 95% CI 1.45, 158.32), similar to the nondiabetic population. Neither cystitis nor pyelonephritis was associated with duration of diabetes, hemoglobin A1c, retinopathy, neuropathy, nephropathy, vascular complications, or glycemic therapy. A prospective study from the Netherlands also reported that only sexual intercourse was associated with symptomatic infection in women with type 1 diabetes (relative risk [RR] 3.6; P 5 .004), whereas asymptomatic bacteriuria was the only significant association for type 2 diabetes (RR 1.65; 95% CI 1.02, 2.67).5 Another prospective study enrolling women in a US health maintenance organization reported increased symptomatic urinary infection in postmenopausal women with diabetes (OR 2.2; 95% CI 1.6–3.0) for subjects receiving oral diabetes medication or insulin.9 A retrospective record review of patients attending primary care practices in the Netherlands reported recurrent urinary infection was increased for women with diabetes (OR 2.0; 95% CI 1.4–2.9).4 The increased risk was independently associated with type 2 diabetes, diabetes of 5 or more years’ duration, receiving oral or insulin therapy, or retinopathy. Hemoglobin A1c was not a risk factor. Studies that have used administrative databases or retrospective record
Table 1 Variables associated with symptomatic urinary tract infection or asymptomatic bacteriuria in women with Type 2 diabetes Risk Factors for Infection (Ref.) Symptomatic4,5
Asymptomatic6
Not diabetes associated
Age
Age
Diabetes associated
Retinopathy Oral hypoglycemic or insulin therapy Diabetes 5 y
Any long-term complication Heart disease Duration of diabetes Oral hypoglycemic therapy
Urinary Tract Infections in Special Populations
review, however, may overestimate the frequency of urinary infection in diabetic women, because these patients are more likely to seek medical care than nondiabetic persons. Diabetes is a common risk factor associated with more severe presentations of urinary tract infection. A case series of 65 consecutive patients with renal or perinephric abscesses reported 28% of subjects had diabetes mellitus.10 In other case series, 67% of patients presenting with emphysematous cystitis were diabetic,11,12 and 62% presented with emphysematous pyelonephritis.11 Diabetes mellitus has not, however, been identified as a risk factor for complications of severe sepsis or septic shock in patients with urosepsis.13 The prevalence of asymptomatic bacteriuria is increased for diabetic women but not diabetic men, with reported rates of 5% to 25% for women and 3% for men.6 Asymptomatic bacteriuria correlates with duration of diabetes and long-term complications of diabetes, but not with parameters of current metabolic control (see Table 1). Management Diagnosis
The clinical and microbiologic diagnosis of urinary tract infection in diabetic populations is similar to other patients with complicated infection. Patients with neuropathy may have impaired bladder sensation, which obscures some clinical symptoms of lower tract infection. Diabetic women with pyelonephritis are more likely to have bilateral renal involvement and bacteremia. In a group of elderly Greek patients hospitalized with pyelonephritis, bacteremia was identified in 30.7% with and 11% without diabetes.14 The types of infecting organisms are also similar for diabetic and nondiabetic patients.15,16 Escherichia coli is the single most common organism. Strains isolated from women with or without diabetes have similar virulence characteristics.3 Treatment
Diabetic women with a functionally normal genitourinary tract are managed similarly to other women with uncomplicated urinary infection. Cystitis should be treated with short-course therapy and pyelonephritis for 7 to 14 days.17 Complicated infections may require more prolonged antimicrobial therapy.16 Imaging is indicated for severe clinical presentations, failure to respond to therapy, or early symptomatic recurrence following discontinuation of antimicrobial therapy. Emphysematous pyelonephritis is managed with antimicrobial therapy and initial percutaneous drainage; delayed nephrectomy, where necessary, is performed once the patient is stable.11 A perinephric abscess usually requires percutaneous or open drainage. Small renal abscesses of less than 3 cm in diameter are treated conservatively with antimicrobial therapy alone and continued until the abscess has resolved on follow-up imaging.10 This treatment may require prolonged therapy for several weeks to months. Outcomes
Hospitalized elderly Greek diabetic patients with acute pyelonephritis had a longer duration of fever (median 4.5 vs 2.5 days, P
