Abstract

A 75-year-old man underwent low anterior resection for sigmoid colon cancer under general anesthesia. Comprehensive preoperative surveys showed normal cardiopulmonary functions. In spite of adequate fluid administration, persistent intraoperative hypotension required vasopressors to maintain an acceptable blood pressure. Although the patient was fully awakened from anesthesia with adequate oxygenation, repeated attempts to wean him from the ventilator failed in the postanesthesia intensive care unit, despite recovery from the effect of muscle relaxant. Low plasma cortisol level was found in a series of laboratory analyses. After supplementation with corticosteroid, the patient was successfully weaned from the mechanical ventilator without any sequelae.

Keywords

corticosteroid insufficiency; steroids; ventilator weaning;

1. Introduction

Corticosteroid insufficiency (CI) can develop in acutely ill patients and may be a factor responsible for failure of weaning from mechanical ventilation after surgery.¹ The etiology of CI and its causal relation with failure of weaning from ventilation are unclear. If the diagnosis of CI is confirmed or highly suspected following laboratory assay, steroid replacement should be instituted until recovery is well settled. The patient under discussion was scheduled for anterior resection of sigmoid colon cancer under general anesthesia, and difficult weaning from mechanical ventilation and low cortisol level were noted. Weaning was successful after steroid replacement therapy.

2. Case report

A 75-year-old man with a height of 182 cm and weight of 98 kg suffered from difficulty in defecation and hematochezia for several months. Colonoscopy disclosed a large ulcerative tumor in the sigmoid colon, so he was scheduled to receive low anterior resection of the tumor under general anesthesia. He had a history of hypertension for >10 years and was regularly treated with atenolol (Tenormin) 100 mg per day. He denied taking Chinese herbal medicine or steroid medication in the past. Comprehensive preoperative evaluation was unremarkable, which included electrocardiogram, ejection fraction (EF) (left ventricular EF: right ventricular EF = 66%: 56%) and ventricular wall motions,chest X-ray, pulmonary function test (forced expiratory volume in one second: forced vital capacity = 2.69 L: 3.67 L = 73%), and laboratory examinations of blood.

Upon arrival at the operating room, his vital signs were: blood pressure (BP) 160/95 mmHg, heart rate (HR) 70 beats/min, and body temperature 36.5°C. Oral clonidine (150 μg) was given as premedication for forestalling hypertension. Standard intraoperative physiological surveying was applied. Anesthesia was induced with 0.5 mg intravenous atropine, 150 μg fentanyl, 500 mg thiopentone and 70 mg rocuronium. The trachea was intubated atraumatically with a 7.5-mm cuffed endotracheal tube. The correct placement of the endotracheal tube was verified by bilateral auscultation of lungs and curve analysis of expiratory CO₂.

No sooner had the endotracheal tube been secured than the patient's BP suddenly fell to 65/35 mmHg, and bradycardia was also noted, and end-tidal CO₂ and O₂ saturation were lower despite manual ventilation with 100% oxygen. Resuscitation was initiated with 24 mg intravenous ephedrine in three separate doses and fluid administration. Hydrocortisone (100 mg) was given intravenously for fear of anaphylactic reaction. Arterial and central venous lines were set up immediately for monitoring. Initial arterial blood gas analysis revealed pH = 7.386, arterial O₂ tension (PaO₂) = 116.6 mmHg, arterial CO₂ tension (PaCO₂) = 44.6 mmHg, base excess = 1.7 mmol/L, [HCO₃⁻] = 25.9 mmol/L, hemoglobin = 13.1 g/dL, [Na⁺] = 140 mmol/L, [K⁺] = 2.8 mmol/L, glucose = 171 mg/dL under ventilator setting at tidal volume 600 mL and respiratory rate at 10 cycles per minute.

During the 6-hour operation, anesthesia was maintained with 1–2% isoflurane, with mean BP maintained in the range 55–100 mmHg, aided by continuous infusion of dopamine at 3–6 μg/kg/min for the latter 3 hours of surgery. Total consumption of rocuronium was 170 mg. The estimated blood loss and total urinary output were 300 mL and 1150 mL, respectively. Fluid replacement consisted of 4.2 L lactated Ringer's solution and 1 L colloid solution.

After the operation, the patient was sent to the postanesthesia intensive care unit (PACU) where he remained intubated for mechanical ventilation. About 2 hours after admission to the PACU, the patient became fully awake with adequate muscle strength. His BP, HR and body temperature were 150/75 mmHg, 60 beats/min, and 35.9°C, respectively. At this juncture, blood gas analysis showed pH = 7.36, PaO₂ = 97.4 mmHg, PaCO₂ = 34.3 mmHg, base excess = −6.4 mmol/L, [HCO₃⁻] = 19.2 mmol/L. A spontaneous breathing trial was performed for preparation of weaning from the ventilator. After 30 minutes evaluation, the trial failed because of hemodynamic instability (BP and HR elevated to >20% of baseline level) and labored breathing. Three repeated trials were taken within the next 8 hours after the operation, but all failed. Another spontaneous breathing trial was repeated in the second postoperative day. In spite of adequate muscle strength and normal plasma electrolyte concentrations, the test failed again. At this moment, plasma thyroid hormones and cortisol level were measured. Levels of tri-iodothyronine, free thyroxine, thyroid stimulating hormone and adrenocorticotropic hormone (ACTH) were all within normal limits, but serum cortisol was far below normal level (cortisol = 1.47 μg/dL). Supplementation of cortisol with intravenous 100 mg hydrocortisone every 8 hours was started on postoperative day 3. On postoperative day 4, the parameters relative to the criteria of weaning from the ventilator positively indicated that the patient could be suspended from mechanical ventilation (Maximal inspiratory pressure: −54 cmH₂O, Maximal expiratory pressure: 56 cmH₂O, Respiratory rate (RR): 29/min, Tidal volume (V_T): 346 mL, Minute ventilation: 10.05 L/min, Rapid shallow breathing index (RSI): 83.8). The patient was successfully weaned from the ventilator on postoperative day 5 when the serum cortisol level remained at around 11.0 μg/dL. He was discharged a few days later without any sequelae and the serum cortisol level was 27.1 μg/dL at the time of discharge.

3. Discussion

We have reported a case of difficulty in weaning from mechanical ventilation due to an uncommon condition of CI. A comprehensive preoperative survey indicated that cardiopulmonary function was normal apart from a history of hypertension for >10 years, which was under regular treatment with atenolol at 100 mg per day. Acute hypotension was noted immediately after induction. At first, we considered that hypotension might have been caused by dehydration due to fasting for >8 hours. The hypotension was not restored although 1,2 L crystalloid and 1 L colloid supplement were given within 2 hours. Continuous infusion of dopamine was required to keep adequate mean BP in the ensuing hours. However, inotropes were no longer needed shortly after the patient became fully awake in the PACU. Therefore, dehydration seemed not to be the most important cause of his hypotension.

The difficulty in weaning from the ventilator 2 days after surgery may have been caused by some insidious disorder such as hypothyroidism or hypocorticoidism. Hypothyroidism was excluded by the normal serum thyroid stimulating hormone, tri-iodothyronine and thyroxine concentrations. However, low serum cortisol was found. During surgery, serum cortisol concentration should increase at the time of incision and reach a maximum at emergence from anesthesia and the early recovery period.² Marik et al have proposed a minimal random cortisol concentration of <25 μg/dL in a highly stressed patient for the diagnosis of adrenal insufficiency.³ Our patient's random cortisol level was 1.47 μg/dL after surgery. A standard (250 μg) rapid cosyntropin test was organized for him after he had been successfully weaned from the mechanical ventilator. His normal response (increase in serum cortisol >9 μg/dL) implied that his adrenal reserve was not impaired.⁴ Hypothalamus–pituitary–adrenal (HPA) and autonomic dysfunction could be the most likely culprit of CI in our patient.

The HPA axis is highly activated during stressful conditions as demonstrated by elevated plasma ACTH levels, cortisol secretion and free cortisol levels. Old age is one of the factors for suppression of the HPA axis. In our patient, the ACTH level was not activated by surgery and low serum cortisol level suggested impairment of HPA function. Our patient might have had a conditionally inadequate corticosteroid response such as in the condition of critical-illness-related corticosteroid insufficiency (CIRCI). CIRCI usually occurs in patients with septic shock or undergoing major surgery.⁵ Our patient fulfilled the common manifestations of CIRCI, that is, hypotension refractory to fluid challenge, and a requirement for vasopressors.

Administration of β-blockers may lead to various degrees of autonomic dysfunction and impairment of the HPA axis as well.⁶ Besides advanced age, chronic use of atenolol might have been the other cause of HPA suppression in our patient.

One may attribute HPA suppression to clonidine. It is an α-2 agonist and may affect autonomic function and sympathetic outflow. However, a previous study has confirmed that premedication with oral clonidine might positively suppress sympathetic neurotransmitters, including norepinephrine and epinephrine, but does not suppress cortisol secretion or production of plasma ACTH.⁷ Therefore, the CI in our patient was not likely caused by clonidine.

The factors causing difficult weaning from mechanical ventilation include imbalance between respiratory muscle strength and load, and weakened intensity of respiratory drive.⁸ The causes of reduced respiratory muscle efficiency include central inhibition, bone marrow diseases, phrenic nerve injury, severe diaphragmatic dysfunction, neuropathies, and neuromuscular junction impairment. The causes of increased ventilator demand include increased central respiratory drive, sepsis, fever, pain, increased respiratory dead space. The principal causes of mechanical ventilation dependence include central nervous system damage, metabolic disorders, endocrine dysfunction, electrolyte disturbances, malnutrition, and anemia. Our patient was somewhat healthy with normal cardiac and pulmonary functions before operation, and had no neuropathies or metabolic disorders in the past. After recovery in the PACU, he was clearly conscious without fever, signs of sepsis, and phrenic nerve injury; his electrolytes and thyroid function were in the normal range apart from the cortisol level, which was lower than the normal range.

Huang et al have found that hydrocortisone therapy for ICU patients with adrenal insufficiency results in improvement in the rate of successful weaning.⁴ They have also found that the percentage of successful ventilator weaning is even higher than that in patients with adequate adrenal function; needless to say, those with adrenal insufficiency without treatment. They have speculated that supplementation with hydrocortisone increases weaning success and improves hemodynamic stability.

In conclusion, CI can develop in acutely ill patients and may be a factor causing weaning failure after surgery. The mechanisms that usher in CI are not yet well defined. We should consider the possibility of CI in cases of unexplained weaning failure, especially in patients with prior glucocorticoid therapy or with unexplained hypotension during a previous operation.