Abstract

d-Lactic acidosis is a rare form of anion gap acidosis and requires a high index of suspicion for appropriate diagnosis and treatment. We report a rare case of short bowel syndrome with recurrent episodes of antibiotic-induced d-lactic acidosis with the presentation of transient severe left ventricular failure. The patient's cardiac function returned to normal with the resolution of acidosis under conservative management and a low carbohydrate diet. d-Lactic acidosis has been known to be associated with only neurological symptoms, and cardiac failure has not been previously reported.

Keywords

acidosis, lactic; heart failure; short bowel syndrome;

1. Introduction

d-Lactic acidosis is a rare form of anion gap acidosis and requires a high index of suspicion even when it presents with classical clinical picture. The problem is further compounded by the fact that all modern arterial blood gas (ABG) analyzers can only measure l-lactate. Thus, in d-lactic acidosis, lactate values as measured will show normal results unless specific tests for d-lactate are ordered. The present case represents a rare scenario, where the episodes of d-lactic acidosis were associated with acute reversible left ventricular failure in a patient with no previous cardiac comorbidities. Increased lactate levels due either to increased production or abnormal metabolism are well known to selectively affect left ventricular functions in many clinical identities. MELAS(mitochondrial encephalopathy, lactacidosis, and stroke syndrome) shows predilection for congestive failure commonly due to dysfunctions of the left heart.¹ Metformin-related metabolic lactacidosis is also known to cause higher incidence of left heart dysfunction progressing to congestive failure.²Increased lactate levels in sepsis have proved to be contributory to left heart dysfunction, highlighting the sensitivity of the left heart to increased lactate levels.³ All previous reports of d-lactic acidosis correlate with the symptomatology of the nervous system, and cardiac involvement has not been previously reported. We discuss the possible causes of precipitating acidotic episodes and sequentially rule out other possibilities leading to acute heart failure in our patient.

2. Case report

A 37-year-old male came to our emergency unit with a fever and productive cough. Upon investigation, his CECT (contrast enhanced computed tomography) of the thorax showed (right-upper zone) lung abscess with pyopneumothorax. He was admitted to the ward, and an intercostal drain (ICD) was inserted. His sputum and ICD samples were both positive for Escherichia coli, for which culture sensitivity-based antibiotics were started. Sputum and ICD discharge tested negative for acid fast bacilli and were also negative for tuberculosis. He sustained deep vein thrombosis (DVT) of the right lower limb 1 year ago during the postoperative period of a bowel resection for ileac tuberculosis. Records showed that he had undergone significant small bowel (jejuno–ileal) resection for active tuberculosis, leaving behind the duodenum and the reservable proximal jejunum only. Prophylactically, he had been on oral warfarin since then for DVT prevention. A Doppler scan of the lower limb showed partial recanalization of the deep veins, and echocardiography revealed normal biventricular function (left ventricle ejection fraction of 55%). He had no history of cardiovascular comorbidities and was ambulatory prior to this episode.

After 1 week's admission, upon resolution of the above symptoms, discharge from the hospital was planned. However, he developed a sudden onset of breathlessness with a decrease of consciousness. His blood pressure fell to 72/56 mmHg with tachycardia (heart rate around 130/min). He was transferred to the intensive care unit (ICU), where urgent tracheal intubation was performed and mechanical ventilatory support was initiated. Dopamine and noradrenaline infusions were started to maintain his blood pressure. His ABG analysis showed metabolic acidosis (pH = 7.1) with high anion gap (24 mEq/L). His blood sugar was within the normal range and urine was negative for ketones. Pulmonary embolism (PE) was suspected because of his clinical presentation and history of DVT. Bedside echocardiography revealed global left ventricle hypokinesia with a left ventricular ejection fraction (LVEF) of 15–20% and preserved right ventricular function (RV) functions. This was a sharp contrast to the echocardiography done 3 days previously, which showed an LVEF of 55%. The possibility of PE leading to hemodynamic instability without altering RV function seemed unlikely. A computed tomography (CT) angiogram was done that ruled out the diagnosis of PE. However, low-molecular-weight heparin and conservative management were started. As the acidosis and hypotension were improved, both noradrenaline and dobutamine were tapered off in 48 hours. His follow-up echocardiography on the 4th day showed normal left ventricular function with ejection fraction improving to 55%. Subsequently, he was returned to the ward 5 days after the episode and allowed to resume oral feeds. He was comfortable and could walk around in the ward. The ICD was removed and discharge was planned.

Three days after his discharge from the ICU, he was suddenly seized again by breathlessness, decreased consciousness, and hypotension, identical with the previous event. ABG analysis showed high anion gap acidosis (31 mEq/L), high pCO₂ (63 mmHg), and hypoxia (pO₂ of 54 mmHg). Hemogram and electrolytes were within normal limits. He was again placed on mechanical ventilationand given inotropes. On auscultation, coarse crepitations were heard over all bilateral lung zones. His chest roentogram showed bilateral hilar shadow with cephalization, suggestive of pulmonary edema (a roentogram taken 2 days previously was almost normal). To rule out possible aspiration due to a decreased conscious level, a bronchoscopy was performed that was negative for any evidence of aspiration. Samples sent for d-dimer levels showed the values were within the normal range (180 ng/mL, normal rate being <250 ng/mL), suggesting against acute PE. A repeat bedside echocardiography again showed global hypokinesia of the left ventricle with LVEF dropping to 25% but the RV function was well preserved (Fig. 1). He was in a state of shock, and blood pressure was only maintained with inotropic support.

Download full-size image

Fig. 1. (A) Previous normal echocardiograph and (B) echocardiograph from the second episode of decompensation with ejection fraction of 25%.

Acidosis-related transient reversible cardiomyopathy was considered. Electrocardiogram only showed inversions of T wave in V1–V4, and no ST changes were seen. Troponin T (qualitative and quantitative) and creatine phosphokinase-MB were negative serially; however, serum pre-pro BNP (brain natriuretic peptide) levels were elevated. His condition was improved with conservative management, pulmonary edema resolved, and inotropic support was tapered off in 3 days. During this episode, samples were sent for evaluation of d-lactate levels because of a suspicious association with short bowel syndrome that accounted for the unexplained acidosis. d-Lactate levels were found to be significantly elevated (5.18 mmol/L, normal <0.2 mmol/L), which signified the causes of the anion gap acidosis and episodic neurological deterioration. Stool cultures were also analyzed and showed the growth of Lactobacillus acidophilus, supporting the diagnosis of d-lactic acidosis of gut origin. Once he was asymptomatic again, he was prescribed oral neomycinalong with a low carbohydrate diet. On resolution of symptoms on the 5th day of the second episode, he was transferred again to the regular ward, and since then no such episodes have occurred.

3. Discussion

The present case placed our ICU in a diagnostic dilemma. The multiple comorbidities in our patient confounded as possible causes of episodes of hemodynamic instability; however, none was altogether able to explain all the clinical consequences seen. The transient nature of the left ventricular failure was the most unprecedented feature, and the unexpected onset and offset of symptomatology added to the complexity of this case. Another aspect that could not be answered in the first episode was whether acidosis was preceded by heart failure or vice versa. Eventually, a thromboembolic event or a vasospastic failure of perfusion of the heart could lead to such a sudden reversible failure,⁴ but both of these possibilities were also negated by the normal troponin levels.

Short bowel syndrome is known to be associated d-lactic acidosis.5, 6, 7Sparing of the colon and loss of the small bowel by resection are prerequisites in considering such a possibility.7, 8 Episodes of d-lactic acidosis are triggered by the use of antibiotics against Gram-negative bacteria, ultimately leading to the natural selection of colonic Gram-positive bacteria.⁹ With an intact small bowel, these bacteria are not subject to the influence of the high-undigested carbohydrates; however, with a major small intestine resection, these carbohydrates would produce d-lactic acid upon reaching the colon. Absorption of d-lactic acid could induce anion gap acidosis and central nervous system symptoms. Major ileum resection for intestinal Koch's leading to short bowel syndrome and lung abscess with E. coli led to the use of antibiotics against Gram-negative selecting d-lactate producing bacteria in the colon. An improvement of his general condition resulted in triggering the episode of d-lactic acidosis. This was, however, not considered at the time of first episode.

With the patient's history of DVT and decreased ambulation due to illness, the diagnosis of PE seemed to be a rational initial conclusion. In a hemodynamically unstable patient in a critical care unit, the use of bedside echocardiography is the most convenient mode of diagnosing major PE.¹⁰ In heart failure due to PE, a major strain occurs in the right heart leading to a decrease in preload to the left heart, thus causing circulatory failure.⁸ The absence of involvement of the right ventricle and isolated severe left ventricular dysfunction were against the possibility of PE leading to the collapse. This was further ruled out by CT angiogram after the initial hemodynamic stabilization. During this episode, no specific cause of mental deterioration could be found, and thus this was attributed to acidosis. The anion gap of 24 mEq/L on ABG also remained partially unexplained, and analysis of the blood sample showed a lactate of 4.2 mmol/L. On conservative management, the patient became almost asymptomatic in the next 72 hours. Echocardiography that was repeated on the 4th day showed that the left ventricular function recovered completely with an ejection fraction of 55%. In a literature search, such a rapid recovery from heart failure could only be related to Takotsubo cardiomyopathy. The diagnostic criteria include: (1) transient hypokinesis, akinesis, or dyskinesis of left ventricle frequently, but not always, a stressful trigger; (2) absence of obstructive coronary disease; (3) electrocardiogram abnormalities (ST-segment elevation and/or T-wave inversion) or modest elevation in cardiac troponin; and (4) absence of pheochromocytoma/myocarditis.¹¹ All these factors coincided with our patient's condition, but such a rapid recovery is unreported in Takotsubo cardiomyopathy. The possibility of acute coronary vasospasm also seemed low because both qualitative and quantitative tests for troponin were serially negative.

The second episode occurred when the patient resumed oral diet. With no other inciting factor, we considered the possibility of d-lactic acidosis, and the elevated levels confirmed the cause of the episodic acidosis. This also explained the altered neurological status in both episodes.¹² A CT scan of the head showed no abnormality, which also supported the possibility of a metabolic cause. The brain, unlike other body organs, cannot metabolize d-lactate because of the absence of d-2-hydroxyacid dehydrogenase and the accumulated d-lactate causes symptomatology. l-Lactate is known to alter the energy dynamics of the heart, more so for the left ventricle,3, 13 which consumes significantly higher energy than the right ventricle. One of the suggested mechanisms is associated with intracellular (cardiac myocyte) acidosis caused by accumulated lactates. d-Lactate and l-lactate have similar physiochemical properties and thus are capable of causing intracellular acidosis. However, the exact mechanisms are not yet clearly elucidated and should be subject to future research.

The altered sensorium was suspected to have caused aspiration; however, this was ruled out by bronchoscopy. Elevated pre-pro BNP levels indicated cardiac failure leading to pulmonary edema.¹⁴ Takotsubo cardiomyopathy has been reported with many clinical stressors; however, the present report is the first one to report with d-lactic acidosis. Once the patient recovered, he was given a low carbohydrate diet along with oral neomycin, which helps to eliminate colonization of L. acidophilus.¹² Thereafter, he remained asymptomatic and was discharged after 1 week's observation.