Escherichia coli is responsible for 80% or more of 16 cases of acute nonobstructive pyelonephritis (Table 23.1). Strains isolated from renal infection belong to a restricted number of O:K:H serotypes characterized by an array of virulence factors that include toxins, such as hemolysin; iron binding proteins, such as aerobactin; and specific adhesin proteins that bind to receptors on uroepithelial cells.¹⁷,¹⁸ These and other virulence genes are often clustered within the genome in pathogenicity islands.¹⁷ Over 95% of strains of E. coli isolated from patients with acute nonobstructive pyelonephritis contain genes for the pap G class II allele of the P fimbria adhesin.¹⁷ The P fimbriae family is one of the mannose resistant adhesins, with a binding specificity for the gal (α1-4) gal-β disaccharide galabiose, which mediates adherence to uroepithelial cells. Specific clonal groups of E. coli, some of which are multidrug resistant, have caused global
outbreaks of uncomplicated urinary infection, including pyelonephritis.¹⁹ For instance, Prats and colleagues²⁰ described a uropathogenic clone (E. coli 015:K52:H1) isolated during a 1-year period in Barcelona, Spain, which was overrepresented in patients with acute pyelonephritis.

Other gram-negative rods, including Klebsiella spp., Proteus mirabilis, and Enterobacter species, are isolated in a few patients with community-acquired renal infections but are much more common in patients with complicated infections. P. mirabilis accounts for more than 40% of infections in infant boys.²¹ P. mirabilis is particularly significant as a renal pathogen because of its propensity to promote struvite calculi.¹⁸ Coagulase-negative staphylococci and Enterococcus faecalis each cause 2% to 3% of invasive renal infections. The latter is a more important pathogen in elderly men. Group B streptococci are isolated in less than 1% of urinary infections, but appear to have a propensity for patients with diabetes and pregnant women.²² Group A streptococci, Streptococcus pneumoniae, and Neisseria spp. are rare upper tract pathogens. Some patients with Staphylococcus aureus bacteremia from other sites will have S. aureus bacteriuria, with or without associated renal infection.²³ Although Staphylococcus saprophyticus is an important cause of acute cystitis in women, its role in invasive upper tract infections is uncertain. Both Mycoplasma hominis and Ureaplasma urealyticum have infrequently been isolated as the sole pathogen in patients with classic acute pyelonephritis; increases in specific antibody titers to these agents support a role for infection.²⁴

Some relatively uncommon organisms isolated include Leptospira spp., Brucella spp., and Salmonella spp. A leptospiral infection usually involves the kidneys and has pathologic changes of interstitial inflammation, hemorrhage, and tubuloepithelial damage.²⁵ The pathogenesis of these lesions is attributed to leptospiral proteins together with hypotension, hypovolemia, and hyperbilirubinemia. Although renal insufficiency is common, localized renal symptoms are unusual; plasma creatinine and blood urea nitrogen (BUN) usually normalize in the second week of illness. A culture of urine on special leptospiral media is usually positive during the acute illness; a polymerase chain reaction or serology are also useful diagnostic tests, if available. Brucella spp. infection may, rarely, be associated with bacterial pyelonephritis and may also present as glomerulonephritis or interstitial nephritis.²⁶ Salmonella spp. are also a rare cause of pyelonephritis, although renal dysfunction is reported to occur in up to 36% of infected adults, usually due to dehydration and rhabdomyolysis.²⁷,²⁸

The microbiology of pyelonephritis in patients with a complicated urinary infection, including catheter-associated hospital-acquired urinary infection, is substantially different. E. coli, usually arising from the patient’s own gastrointestinal tract, remains the most common urinary pathogen, but many other species are frequently isolated, and bacteria are likely to be of increased resistance. The spectrum of infecting organisms in a patient will be influenced by exposures to the health care environment as well as current or recent antimicrobial exposure. E. coli isolated from pyelonephritis in patients with complicated urinary infection have a much lower prevalence of potential virulence factors, including the P fimbria, consistent with host rather than organism factors being the principal determinants of infection.²⁹,³⁰,³¹ More resistant gram-negative rods, including Pseudomonas aeruginosa and Serratia marcescens, account for 10% to 15% of hospital-acquired invasive upper tract infections in some reports, and may occur in outbreaks.³² Corynebacterium group D2 has been identified as a unique etiologic agent of nosocomial urinary infection, particularly in catheterized patients.³³ These organisms are urease producers and may be isolated from persistent infections including bladder and renal calculi, pyelonephritis, and bacteremia. They are slow growing and sometimes missed if routine cultures are discarded after 24 hours.

Organisms that are unable to use urine as a nutrient source only rarely cause pyelonephritis. These include most species of obligate anaerobes, for which the relatively high oxygen tension in normal urine also likely inhibits growth. In a prospective study of 5,781 urine specimens, Segura et al.³⁴ identified only 10 patients with positive urine Gram stains and negative aerobic cultures from which anaerobic bacteria were isolated— an overall prevalence of 1.2% of bacteriuric specimens. All but 1 of these 10 patients had complicated urologic problems. Uropathogens surviving in the kidney without cell walls, also called L forms or protoplasts, have been suggested to contribute to relapsing pyelonephritis by enabling organisms to persist and cause chronic disease in the hypertonic environment of the renal medulla.³⁵ However, studies have not yet confirmed a role for these bacterial forms in urinary infections.

Host factors contributing to pyelonephritis and the host immune and inflammatory response are described in detail in Chapter 21. Behavioral factors associated with acute nonobstructive pyelonephritis in premenopausal women are similar to those for acute uncomplicated cystitis—most importantly sexual intercourse and spermicide use for birth control.¹² There is also a genetic predisposition, as evidenced by a twofold to sixfold increased prevalence of urinary infections in the mothers and female siblings of girls and women with recurrent urinary infections.³⁶ Women who are nonsecretors of the blood group substance have an increased risk of recurrent acute uncomplicated urinary infections,³⁷ and selected blood group antigens are associated with an increased frequency of urinary infections in girls without vesicoureteral reflux.²¹ Epithelial cell receptors necessary for E. coli binding are glycolipids of the globoseries, the antigens of the P blood group system. Recent studies exploring polymorphisms of effector molecules of the innate immune response have described other potential genetic determinants. In children without vesicoureteral reflux, pyelonephritis is reported to be associated with polymorphisms and mutations of the CXCR1 receptors for interleukin (IL)-8 and GCP-2 in some,³⁸,³⁹ but not all studies.⁴⁰ A single nucleotide polymorphism for IL-8 was associated with severity of pyelonephritis among
children.⁴⁰ Genetic variation in the Toll-like receptor (TLR) promoter TLR-4 may also influence susceptibility to pyelonephritis in children,⁴¹ whereas a TLR-1 receptor polymorphism has been reported to be associated with protection from pyelonephritis in women.⁴²

Individuals with structural or functional abnormalities of the urinary tract have a greatly increased risk of pyelonephritis, apparently independent of behavioral or genetic factors (Table 23.2). Obstruction at the level of the kidney or ureter may directly inhibit urine flow from the upper tract, allowing bacteria to establish infection behind the obstruction. In addition, voiding abnormalities, such as neurogenic bladders, are frequently associated with reflux, which promotes upper tract infections.

Acute pyelonephritis is characterized by an intense local and systemic inflammatory and immune response.¹⁷ Following the stimulation of epithelial receptors within the urinary tract by bacteria or bacterial products, there is activation of the innate immune response through TLRs. The Pap G adhesin appears to be important in stimulating the epithelial cytokine response.⁴³ Pyuria occurs early, and urinary and serum cytokines and chemokines including IL-6, IL-8, and others are elevated.⁴²,⁴³,⁴⁴,⁴⁵ Alterations in the expression of antimicrobial peptides may be another element of the innate response, but there is a limited evaluation to date of the role of these molecules within the urinary tract.⁴⁶ The intensity of the inflammatory response correlates with the severity of symptoms. Pyelonephritis is associated with a greater inflammatory response than a lower tract infection, and urinary cytokines have increased levels in symptomatic compared with asymptomatic UTIs. This inflammatory response resolves promptly with the institution of effective antimicrobial therapy.⁴⁷ Renal infection is also associated with both a local and a systemic humoral immune response. An immunoglobulin (Ig)M antibody predominates with the initial infection but subsequent infections are dominated by an IgG response.¹⁷,¹⁸ The primary antibody response develops about 7 to 10 days following the initial infection. Local urine antibody includes both an IgA and an IgG response to the infecting bacteria. Whether this humoral immune response has any protective role for subsequent infection remains controversial.

A renal biopsy is contraindicated for patients with acute pyelonephritis. When pathology specimens are available, the histologic hallmarks of acute pyelonephritis include abscess formation and edema in the renal parenchyma with an accumulation of polymorphonuclear leukocytes in and around the tubules. Bacteria are often demonstrable in the foci of acute renal suppuration. In general, glomeruli are spared, although small abscesses may surround them. Areas of infection are characteristically wedge-shaped with the apex in the medulla resembling an infarct. Although tissue destruction is greater in the cortex than in the medulla, the relative smaller size of the medulla means the inflammatory response appears to have a greater effect on medullary anatomy and function. The distribution of wedge-shaped areas of suppuration is characteristically focal, usually corresponding to renal lobes, and sharply demarcated from areas of uninvolved renal parenchyma. In adults, the kidney is unlikely to be uniformly affected unless concomitant obstruction is present.

Chronic pyelonephritis is a focal parenchymal disease with associated changes of the renal collecting system owing to inflammation and deformation. Fibrosis with atrophy of overlying renal tissue leads to surface depression or scars. Neighboring unaffected renal tissue often undergoes hypertrophy and may appear to be a mass lesion. A sharply defined border between normal and diseased tissue is characteristic of chronic pyelonephritis. The capsule is adherent and the cortical surface is irregular. Calyceal clubbing results from a papillary retraction into the scar. Dilatation, muscular hypertrophy, and fibrotic inflammation causing a thickening of the calyceal system all occur to a variable extent. The two kidneys are usually markedly asymmetrically involved, whereas other diseases that cause interstitial inflammation usually affect both kidneys equally.

The histology of chronic pyelonephritis is the pathology of interstitial or tubulointerstitial nephritis and is characterized by a pleomorphic infiltrate of lymphocytes, plasma cells, and macrophages in the interstitium of the kidney. Polymorphonuclear leukocytes and eosinophils may also be present. The vessels in zones of the normal kidney may be normal or may demonstrate hyaline intimal changes. Histologic features of acute and chronic pyelonephritis can overlap. The relative degrees of edema and fibrosis rather than the interstitial cellular response are the most useful criteria to delineate these entities. Previously, many end-stage kidneys were referred to pathologically by the term chronic pyelonephritis. This is now recognized as being a common end stage for many renal diseases, and is seldom attributable to infection.

Bacteriuria occurs in 0.7% of full-term infants, and clinically evident urinary infections occur in 0.3%.²¹ Males predominate in the first 3 months of life, and account for almost 80% of neonatal urinary infections. Most of these infections appear to involve the upper tracts. In a retrospective review of 11,655 children born at a Stockholm hospital between January 1, 1979 and June 30, 1982, the annual incidence of pyelonephritis during the first 2 years of life was 34 per 10,000.⁴⁸ Acute pyelonephritis occurred after a mean of 5.6 months following discharge, with a range of 1 week to 17 months; only one infant had an underlying malformation of the urinary tract. A more recent report from a Seattle group health cooperative reported hospitalization rates for pyelonephritis were 17/10,000 for girls and 6/10,000 for boys age 0 to 4 years.⁴⁹ It is estimated 1% to 3% of all girls 1 to 5 years of age experience an episode of pyelonephritis.²¹

About 250,000 episodes of acute pyelonephritis with over 160,000 hospitalizations occur each year in adult women in the United States.⁵⁰ The highest rates of acute nonobstructive pyelonephritis are in young women, many of whom also experience acute uncomplicated cystitis (Table 23.3). Ikaheimo and associates⁵¹ reported an incidence of 2.7 episodes of pyelonephritis per 100 patient years during a 1-year follow-up of women in a family practice in Finland who originally presented with an episode of acute cystitis. The ratio of episodes of cystitis to pyelonephritis was 29:1. Stamm and associates⁵² followed 51 American women with recurrent uncomplicated urinary infections for a median of 9 years. The mean infection rate was 2.6 per patient year with the ratio of cystitis to pyelonephritis episodes of 18:1 for women not receiving prophylactic antimicrobials.

Hospitalization for treatment of acute pyelonephritis reported from a Canadian province was 11/10,000 women.⁵⁴ Pyelonephritis accounted for 0.4% of all hospital admissions.
The frequency of hospitalization underestimates the incidence of pyelonephritis because many patients with pyelonephritis, especially healthy young women, are not admitted for treatment. Peak pyelonephritis hospitalization rates occurred in women 20 to 29 years old, and men and women older than 50 years. From 0.3% to 0.7% of all pregnancies required hospitalization for pyelonephritis. Women with diabetes were six to 24 times more likely to be admitted with pyelonephritis than nondiabetic women, stratified by age. An American study reported a similar hospitalization rate of 11.7/10,000 for women,⁵⁰ but hospitalization rates were not increased in patients with diabetes. The highest rates were observed in younger women aged 20 to 39 years, with 15 hospitalizations/10,000, and women over 80 years, with 23.3/10,000. For women enrolled in a group health cooperative in Seattle, estimated annual rates of outpatient and inpatient pyelonephritis were 12 to 13/10,000 and 3 to 4/10,000, and the highest incidence of 18 to 20/10,000 occurred in young women.⁴⁹

Pyelonephritis in men usually occurs in the context of complicated urinary infections, but acute nonobstructive pyelonephritis may rarely occur. Krieger and colleagues⁵⁵ reported an incidence of uncomplicated symptomatic urinary infection of 4.9/10,000 men per year in a 6-year study of male university students, but most of these were likely lower tract infections. Rates of hospitalization for acute pyelonephritis in men, most with complicated urinary infections, were reported to be 3.3/10,000 per year in Canada,⁵⁴ and 2.4/10,000⁵⁰ and 1 to 2/10,000⁴⁹ in the United States. In selected populations with complicated urinary infections, Waites and associates⁵⁶ reported 1.8 episodes per person year of urinary infections presenting with fever and chills, presumably upper tract infections, in 64 spinal cord-injured patients managed with intermittent catheterization or condom drainage. A prospective study of residents in long-term care facilities reported 1.1 episodes of febrile urinary infection per 10,000 resident days.⁵⁷ The rate was 0.8/10,000 resident days for individuals without chronic indwelling catheters, and 4.6/10,000 for those with chronic indwelling catheters.

Asymptomatic bacteriuria is common in many populations,³ and frequently involves the upper urinary tract (Table 23.4). Presumably, factors such as the duration of bacteriuria, infecting organisms, associated medical illnesses, the presence of
vesicoureteral reflux, and urinary obstructions influence the likelihood of a UTI in patients with asymptomatic bacteriuria. At least 50% of institutionalized elderly women with bacteriuria have upper tract localization.⁶³,⁶⁴ Bacteriuria in these populations often persists and remains asymptomatic for months to years. Bacteriuria is likely attributable to the same biologic variables as recurrent uncomplicated or complicated symptomatic infection, but bacterial isolates causing bacteriuria are rarely a direct cause of subsequent symptomatic infection in the absence of uroepithelial trauma or obstruction.

The short-term morbidity of acute pyelonephritis in children may include hospitalization, severe sepsis and septic shock, metastatic infection, and, rarely, acute renal failure. Following the introduction of antimicrobial therapy, the outcome for adequately diagnosed and treated children is excellent. However, there remain concerns about the potential for long-term renal damage following acute pyelonephritis in childhood.

Renal scarring is observed in 10% to 30% of children following acute pyelonephritis.²¹ Established risk factors for the development of renal parenchymal scarring include vesicoureteric reflux, recurrent infection, delayed treatment, and a young age at the time of initial infection.⁶⁵ Parvex et al.⁶⁶ observed 88 scars in 50 children at 6 months after acute pyelonephritis; 3 years later, 27% were unchanged, 63% were partially resolved, and 9% were completely disappeared. The number of scars was the most important variable associated with decreased renal growth. Increased serum and urine markers of the inflammatory response at presentation with acute pyelonephritis, including IL-8,⁶⁷ IL-6,⁶⁸ procalcitonin,⁶⁹ and C-reactive protein,⁶⁹ are associated with subsequent increased occurrence of renal scarring, regardless of vesicoureteral reflux. Recent studies have suggested that cytokine gene polymorphisms may partly explain the differential inflammatory response.⁶⁵,⁷⁰,⁷¹ Studies do not support a role for bacterial virulence factors as predictors of subsequent scarring.⁷²

The relative importance of vesicoureteral reflux and pyelonephritis in the development of renal scars and impaired renal function in childhood remains controversial. It is now accepted that congenital vesicoureteral reflux, primarily occurring in males and associated with higher levels of reflux and renal dysplasia, is most likely to progress to renal failure, regardless of infection.⁷³ Postnatal renal damage may also occur with vesicoureteral reflux associated with an infection, or with acute inflammation from an infection of the renal parenchyma without reflux. Scarring after pyelonephritis in infancy is associated with renal growth arrest in the involved kidney and may be associated with progressive kidney damage.⁶⁷,⁷³ The risk of renal failure is significantly greater with high-grade reflux (grade IV through V) and with multifocal or global scarring. A careful follow-up with an immediate diagnosis and the adequate treatment of all recurrent episodes of infection, particularly during infancy and early childhood, are considered necessary to prevent progressive renal impairment.⁷³

Asymptomatic bacteriuria occurring in girls 5 years or older with normal kidneys on a study entry is benign.⁷⁴ It is not associated with renal scar development regardless of antimicrobial treatment. Thus, programs to screen for bacteriuria in school-age girls and to treat infections, if found, are not worthwhile.

A comprehensive population-based study of the incidence and prevalence of bacteremic acute pyelonephritis from 1977 to 1981 in an urban population of 400,000 reported 22% of community-acquired bacteremias were attributed to invasive urinary infections, with an annual incidence of 15.7 per 100,000.⁷⁸ The attributable mortality for bacteremic urinary infections was 4.8%, but all 15 deaths occurred in patients with a significant underlying illness. During the same period, these investigators observed 1,520 episodes of hospital-acquired bacteremia, of which 221 (14.5%) originated from the urinary tract (71% in catheterized patients), thus yielding a rate of 7.3 per 10,000 hospitalized persons.⁷⁹ The mortality rate attributed directly to infections in these patients with bacteremic nosocomial pyelonephritis, virtually all with complicated urinary infection, was 12.7%. This mortality rate is one-third that of gram-negative bacteremia originating from other sites. A review of 542 episodes of bacteremic gram-negative urinary infection in Olmsted County, Minnesota, from 1998 to 2007 included 57% that were community acquired, 36% that were health care associated, and 7% that were nosocomial.⁸⁰ All cause mortality was 4.9% at 28 days and 15.6% at 1 year. The only independent predictor of increased mortality was increasing age, whereas a lower mortality was associated with community-acquired infections and an isolation of E. coli. In critical care units,
14.6% of patients admitted with septic shock had a urinary source.⁸¹ The 28-day mortality when the urinary tract was the origin of the septic shock was only 18%, compared with 36% for all other sites. Berger et al.⁸² described a case series of 65 patients admitted from 1994 to 2007 at one Australian center who required emergency nephrectomies due to severe urosepsis; the mortality in these individuals was 20%. Thus, mortality attributable to pyelonephritis occurs, but virtually only in patients with complicated infections.

Genitourinary sepsis accompanied by bacteremia may be complicated by metastatic infections at other sites. Siroky and colleagues⁸³ identified 175 patients in whom metastatic infections developed from a primary source in the genitourinary tract. Most patients (86%) were men and the mean age was 57 years. A primary prostatic focus was considered the site for dissemination in many of the men, but in 46 patients the upper urinary tract was the likely source of the bacteremia. One hundred six patients (59%) had infections of the skeletal system; 51 patients (28%) had endocarditis; and 13% involved miscellaneous sites, including the eye and the central nervous system. Almost 70% of the skeletal infections were caused by gram-negative rods, with the vertebral column the most common metastatic site, identified in 83 patients (78%) with skeletal infections. In patients with endocarditis, gram-positive organisms were responsible for two-thirds of the infections, and these patients usually had preexisting heart disease. About one-third of these patients had undergone a manipulation of the upper urinary tract prior to the development of metastatic infections. Underlying host factors that would impair resistance to infections were unusual. In a Danish review of cases of vertebral osteomyelitis from 1978 to 1982, the urinary tract was the most common identified source, with the mean latent period separating an episode of acute urinary infection from the onset of symptoms of vertebral osteomyelitis being 54 days.⁸⁴

Adult onset pyelonephritis rarely contributes to chronic renal failure. Abnormalities previously attributed to pyelonephritis in autopsy studies are now recognized as being an end stage of other processes such as vascular disease, papillary necrosis, or medullary cysts, with little contribution from infections. In one autopsy study of patients with renal failure, 13% of subjects were considered to have pyelonephritis, but all had vesicoureteral reflux, analgesic abuse, nephrolithiasis, or obstruction as underlying contributory factors for pyelonephritis.⁸⁵

The Bristol Pyelonephritis Registry followed 375 women for 1 to 13 years after a clinical diagnosis of recurrent pyelonephritis. Only one patient had a radiographic progression of renal scars.⁸⁶ Other long-term studies that have reported patients with progression to renal failure also invariably identify alternate diagnoses to explain renal functional deterioration in affected patients. Parker and Kunin⁸⁷ retrospectively reviewed 74 cases among 163 women hospitalized for acute pyelonephritis 10 to 20 years previously, in the early antimicrobial era. Continuing clinical illness following the index hospitalization had occurred in more than 40% of patients; 28% had had an operative urologic procedure, and 23% had renal stones. Seventeen percent were bacteriuric at the follow-up examination. One patient died of complications of pyelonephritis, one required a transplantation for end-stage renal disease, and two others had significant renal impairment. Seven patients had undergone unilateral nephrectomy for pyelonephritis. Gower⁸⁸ followed 62 adult women with treated infections and a radiologic diagnosis of chronic pyelonephritis for a mean follow-up of almost 5 years. Serial studies demonstrated radiographic progression in 11 women, but persisting infection and analgesic ingestion contributed to progressive radiologic damage in all these cases. Two of the 36 patients with bilateral pyelonephritis had renal failure or died during the follow-up. Alwall⁸⁹ described a selected series of 29 women with an initial normal-appearing intravenous pyelogram (IVP) who developed contracted kidneys from 1 to 15 years following acute pyelonephritis. Several went on to end-stage renal disease, but for all these cases, analgesic abuse was a concomitant factor that likely accounted for disease progression.

Raz and colleagues⁹⁰ described long-term outcomes for women admitted with acute pyelonephritis to a hospital in Israel between 1982 and 1992. Only 31% of the patients were available for a 10-year follow-up, likely a group representing those with more serious or persistent disease. With technetium ⁹⁹Tcm-labeled dimercaptosuccinic acid (⁹⁹Tcm-DMSA) scanning, 46% of these women had evidence of renal scarring. Pregnancy and hypoalbuminemia at hospitalization for pyelonephritis were independent risk factors correlated with the finding of renal scars at a 10-year follow-up. Although four women with scars had a glomerular filtration rate (GFR) of less than 75 mL per minute, none had developed renal impairment. Despite the high proportion of this selected group of women who had renal scarring, there were no clinically relevant adverse outcomes.

These long-term prospective studies support the observation that recurrent UTIs in adult women usually have a benign natural history. Adults with recurrent UTIs and no other complicating illness seldom experience clinically significant renal damage directly attributable to an infection. Persons with infection and renal impairment invariably have significant underlying urologic abnormalities or associated renal diseases.

The long-term follow-up of cohorts that have enrolled large numbers of women have consistently reported no significant differences in blood pressure between patients with bacteriuria and those without bacteriuria.³ Patients entered into the Bristol Pyelonephritis Registry and followed for up to 13 years developed hypertension at the same rate as the
general population.⁸⁶ Raz et al.⁹⁰ reported a similar frequency of hypertension in women with and without renal scarring 10 to 20 years after hospitalization for acute pyelonephritis. Thus, evidence does not suggest acute or recurrent pyelonephritis contributes directly to development of hypertension.

Acute pyelonephritis occurs in 1% to 2% of all obstetric patients in the absence of screening and treatment programs for bacteriuria.⁹¹ It is the most common medical complication requiring hospitalization during pregnancy. Women with asymptomatic bacteriuria early in pregnancy have a 20- to 30-fold increased risk of acute pyelonephritis in later trimesters.⁹² This is attributed to ureteral dilation and urinary stasis resulting from progesterone-induced smooth muscle relaxation, together with mechanical compression by the enlarging uterus. Acute pyelonephritis in pregnancy occurs primarily in the second and third trimesters. Case series have reported 52% of episodes occurring in the second trimester, 46% in the third, and 2% in the first,⁹³ and 11% occurring in the first trimester and the remainder in the second or third trimester.⁹⁴

Acute pyelonephritis, as with any febrile bacterial illness in late pregnancy, is associated with an increased risk for premature labor, presumably due to systemic inflammation.⁹⁵ The attributable risk of acute pyelonephritis for maternal toxemia, prematurity, and perinatal mortality remains controversial.⁹¹ Sever and associates,⁹⁶ in data collected from more than 55,000 pregnant women, reported a higher incidence of low-birth-weight infants and stillbirths in the 3.5% of women with documented UTIs. Naeye⁹⁷ reported a combined perinatal mortality rate of 42 per 10,000 births in bacteriuric women as opposed to 21 per 10,000 births in nonbacteriuric women. McGrady and colleagues,⁹⁸ using birth certificate data from Washington State, showed that the fetal mortality rate was 2.4 times higher for UTI-associated pregnancies. Romero and colleagues,⁹⁵ in a meta-analysis, documented an increased occurrence of low-birth-weight and preterm delivery with asymptomatic bacteriuria. Smaill (in a Cochrane review)⁹⁹ also showed that antibiotic treatment significantly reduced the risk of low birth weight. Thus, asymptomatic bacteriuria is associated with prematurity and low birth weight, but it is not clear whether acute pyelonephritis is also a risk, or whether any presentation of urinary infection is causative for these outcomes.

Pregnant patients have a reduced GFR following acute pyelonephritis, which reverts to normal within 8 weeks of effective treatment.¹⁰⁰ Long-term follow-up studies of women known to have previously experienced bacteriuria during pregnancy report a benign course for the majority of patients.¹⁰¹,¹⁰² In a follow-up period of 10 to 14 years, almost 40% of 134 women with bacteriuria during pregnancy also had bacteriuria on follow-up cultures. Although pyelography showed renal scarring in 28% of these patients, creatinine clearance was normal for both the bacteriuric and the nonbacteriuric women.¹⁰² Raz et al.⁹⁰

Stay updated, free articles. Join our Telegram channel

Join