Sepsis is defined as suspected or proven infection plus a systemic inflammatory response syndrome (e.g., fever, tachycardia, tachypnea, and leukocytosis). Severe sepsis is defined as sepsis with organ dysfunction (hypotension, hypoxemia, oliguria, metabolic acidosis, thrombocytopenia or obtundation), whereas septic shock is defined as severe sepsis with hypotension, despite adequate fluid resuscitation. Potential causes of severe sepsis or septic shock during pregnancy and the postpartum are listed in Table 8-1 .
|
The presenting signs and symptoms in severe sepsis during pregnancy can be variable depending on the etiology as well as the duration of infection ( Table 8-2 ). The most common presenting symptom is fever (>39.0° C [>102° F]) with or without chills; however, in cases of advanced sepsis the patient can develop hypothermia (temperature <36.0° C [<96° F]) with tachycardia (heart rate [HR] >110 beats per minute [bpm]) and tachypnea (respiratory rate [RR] >24/min). In most cases, the location of pain or tenderness assists in determining the etiology of the underlying infection. For example, patients with pyelonephritis present with flank or back pain and the tenderness localized at the costovertebral angle, whereas those presenting with either cholecystitis or pancreatitis have right upper quadrant abdominal pain and tenderness or generalized abdominal pain. On the other hand, patients with endomyometritis, chorioamnionitis, and septic abortion have lower abdominal and pelvic pain in association with uterine and cervical motion tenderness.
|
The laboratory findings in patients with severe sepsis or septic shock depend on the etiology, the duration of infection, presence of preexisting medical or obstetric disorders, as well as the quality of management used ( Table 8-3 ). It is important to note that traditional laboratory values to define sepsis in nonpregnant women may not apply to pregnant women. The most common laboratory abnormality in patients with septic shock during pregnancy is leukocytosis (usually a white blood count >15,000/mm 3 ); however, in cases of advanced sepsis the patient can develop leukopenia and neutropenia. In addition, patients with viral sepsis usually have leukopenia. Moreover, most patients have abnormal serum creatinine levels (>1 mg/dL).
|
Pregnancies complicated by severe sepsis or septic shock are associated with increased rates of preterm labor, fetal infection, and preterm delivery resulting in increased rates of perinatal morbidity and mortality. These complications are related to prematurity or maternal hypoxemia and acidosis. Although septic shock is rare during pregnancy, its development may result in substantial morbidities and even maternal death ( Table 8-4 ). Indeed, the reported maternal mortality from septic shock ranges from 20% to 50% (with higher rates seen in patients with multiple organ failure).
|
Case 1: Severe Sepsis
A 21-year-old G1P0 at 31 weeks of gestation presents during a routine prenatal office visit. She complains of burning on urination, right flank pain, and nausea and vomiting. A urinalysis with a urine culture is obtained. She refuses antibiotics stating, “It might hurt the baby,” and the physician does not insist on treatment.
The patient presented 48 hours later to the labor and delivery unit with worsening pain and complaints of fever and chills. Clinical evaluation revealed the presence of right costovertebral angle tenderness in association with the following:
- •
Temperature (100.4° F oral)
- •
RR (22/min)
- •
HR (110/min)
- •
Systolic blood pressure (SBP) 90 mm Hg
- •
O 2 saturation on room air 94%
Laboratory evaluation revealed:
- •
White blood cell count (20,400/mm 3 with 15% bands)
- •
Platelet count (160,000/mm 3 )
- •
Serum creatinine (1.3 mg/dL)
- •
Serum lactate (3.1 mmol/L)
The organism on urine culture from her previous office visit returned Escherichia coli; however, antibiotic sensitivities were not yet available. The patient had no known drug allergies and was started on IV ampicillin 2 g every 6 hours. IV fluid of D 5 normal saline (NS) was started at a rate of 125 mL/hr with no initial bolus of fluid. She was also started on 2 L of oxygen/minute by nasal cannula for tachypnea. Fetal heart rate monitoring revealed fetal tachycardia with moderate variability and absent periodic decelerations. Uterine monitoring revealed the presence of contractions (1 every 10 minutes). There was no cervical change or evidence of rupture of membranes on pelvic examination.
Eight hours following admission to the hospital her clinical evaluation changed to the following:
- •
Temperature (102.4° F oral)
- •
RR (60/min)
- •
HR (140/min)
- •
SBP (70 mm Hg)
- •
O 2 saturation on 2 L oxygen by nasal cannula 88%
- •
Urine output for 8 hours (120 mL)
Repeat laboratory evaluation showed the following:
- •
White blood cell count (31,200/mm 3 with 20% bands)
- •
Platelet count (90,000/mm 3 )
- •
Serum creatinine (2.1 mg/dL)
- •
Serum lactate (5.8 mmol/L)
- •
Serum AST (100 IU/L)
Fetal heart rate monitoring continued to reveal fetal tachycardia, but now also with minimal variability, absent accelerations, and no decelerations. Uterine contraction monitoring revealed a contraction frequency of 1 every 6 minutes. Repeat cervical examination was unchanged. Because of her severe hypoxia and tachypnea, an anesthesiologist was paged to intubate the patient, which was accomplished without difficulty. Following intubation, the patient was transferred to the intensive care unit. An immediate chest x-ray noted appropriate placement of the endotracheal tube but lung findings were consistent with acute respiratory distress syndrome (ARDS). A diagnosis of severe sepsis was made and a severe sepsis protocol was initiated that included infectious disease, pulmonary medicine, and pharmacy consultations, as well as the initiation of broad-spectrum antibiotics and central hemodynamic monitoring.
Subsequent management included a change in antibiotic therapy to the third-generation cephalosporin, ceftazidime 1 g IV every 8 hours. Aminoglycosides such as gentamicin could also have been chosen for antibiotic coverage for E. coli but must be used in caution with impaired renal function due to their potential for nephrotoxicity. Management should also include placement of invasive hemodynamic monitoring for fluid management and evaluation of cardiac function as well as acetaminophen therapy to reduce maternal fever to potentially improve fetal tachycardia. The urine culture sensitivity results revealed that the E. coli was resistant to ampicillin but sensitive to ceftazidime. With aggressive therapy the patient had gradual improvement including extubation 48 hours later and becoming afebrile 72 hours after the change in antibiotic therapy. Fetal heart rate monitoring revealed a normal baseline with moderate variability and presence of accelerations. Uterine monitoring revealed no contractions. The patient was discharged home on suppression antibiotics and subsequently had a spontaneous vaginal delivery at term.
Discussion
Early detection of disease process and intervention can improve the outcome and survivability in severe sepsis and septic shock. Early provision of time-sensitive therapies and standardized therapies of best practice have been shown to decrease mortality, hospital cost, and hospital length of stay in randomized studies in complicated patients. This finding underscores the importance of developing and implementing a severe sepsis protocol that includes goal-directed therapy. This requires the involvement of a multidisciplinary approach that includes physicians, nursing, pharmacy staff, and hospital administration.
In this case, the initial antibiotic choice of ampicillin alone was inappropriate because many E. coli isolates in pregnancy are resistant to this agent. The patient should have received a combination of agents such as ampicillin and gentamicin or initial treatment with an alternate agent such as a cephalosporin. In addition, fluid resuscitation should have been more aggressive and used normal saline or lactated Ringer’s solution. In patients presenting with acute pyelonephritis in pregnancy with hypotension, initial fluid resuscitation should be performed initially as rapid infusions of warmed fluids (1 to 2 L over 60 minutes). Subsequent IV infusions are guided by maternal vital signs, pulse oximetry, and urine output to avoid the development of pulmonary edema. Colloids do not appear to be superior to crystalloids but attempts should be made to avoid excess free water (i.e., with use of 0.9% NS).
During the observation period, the patient was having a gradual deterioration in her cardiorespiratory status that culminated with the development of septic shock. Appropriate management should have included the initiation of the severe sepsis protocol. This would have consisted of admission to an intensive care unit or a special obstetric care unit for close and continuous monitoring of vital signs, oxygenation, and fluid intake/output. To achieve adequate organ perfusion and avoid the development of pulmonary edema, fluid resuscitation should be directed by central hemodynamic monitoring with physiologic perfusion endpoints including mean arterial pressure ≥65 mm Hg, central venous pressure 8 to 12 mm Hg or urine output >25 L/hr. This management could have avoided the need for mechanical ventilation and progression to septic shock.
Patients with acute pyelonephritis during pregnancy can develop uterine contractions (with or without cervical change) as a result of release of endotoxins. In general, most patients respond to hydration and the contractions resolve after treatment. As a result, true preterm labor is not common. However, on occasion, a patient develops true preterm labor with cervical dilation, suggesting the need for tocolytic therapy. A major concern with tocolytic therapy in this clinical setting is increasing the risk of pulmonary edema particularly if β-agonists are used. Therefore, in certain instances such as gestational age less than 32 weeks, tocolytic therapy with magnesium sulfate to allow time for corticosteroid administration for fetal benefit may be considered.