Non-diabetic euglycaemic ketoacidosis and rapid weight loss in a post-traumatic surgical patient: is the outré preventable?
- 1 Department of Anaesthesia, Intensive Care and Pain Medicine, Singapore General Hospital, Singapore
- 2 Department of Anaesthesiology, Jurong Health Campus, National University Health System, Singapore
- Correspondence to Dr Dinoo Suran Kirthinanda; dinoo.sn@live.com
Abstract
To highlight the implications of the metabolic stress response and the role of anaesthesia in attenuating its deleterious effects, we present this extremely rare case of non-diabetic euglycaemic ketoacidosis with rapid weight loss in a post-traumatic surgical patient. Ketoacidosis is the accumulation of ketone bodies in blood and is generally associated with relative or absolute insulin deficiency secondary to diabetes mellitus, sodium–glucose cotransporter 2 inhibitors and extensive fasting. The stress of systemic disease, trauma or surgery in such predisposed patients could precipitate ketoacidosis. Our patient developed high anion gap metabolic acidosis intraoperatively due to ketosis, a potentially life-threatening complication, without any predisposing factors as a result of metabolic stress of major trauma and surgery. Aiding the interpretation, he lost 15 kg weight perioperatively, suggesting his body was in a hypercatabolic state. This report emphasises the value of anaesthetic techniques to prevent such rare complications.
Background
A middle-aged man was admitted to the hospital after a high-velocity road traffic accident (RTA). He sustained multiple fractures and lacerations outlined below. He is a smoker (20 pack-years) and non-drinker and has a previous medical history of hypertension, hyperlipidaemia and settled adrenal insufficiency secondary to exogenous steroid consumption from adulterated herbal products 3 years ago. He has since been discharged from Endocrinology based on clinical and hormonal assays. He was not on steroid replacement and did not have Cushingoid features on admission.
As a result of the RTA, he suffered multiple injuries: lacerations over the bilateral lower limbs and right forearm, right comminuted clavicle fracture, right scapular fracture, multiple right 2nd to 12th rib fractures with pulmonary contusion and a small right haemothorax.
Case presentation
This patient underwent bilateral lower limb wound debridement on day 1 of hospital admission under an uneventful central neuraxial block. Postoperatively, the patient was allowed a complete diet. An adequate analgesia regimen was established with satisfactory pain control until the fracture fixation. His oxygenation has been adequate, and he was able to participate in breathing exercises. He has been evaluated by the Endocrine Team and did not require further optimisation prior to the surgery. The patient was kept nil by mouth from midnight of day 4 of injury in preparation for surgery the next morning for the fixation of right-sided rib, clavicle and scapular fractures with insertion of chest tube, bilateral lower limb wound debridement with split skin grafting in a 10-hour emergency surgery under general anaesthesia.
In addition to standard American Society of Anaesthesiology monitoring guidelines, an intra-arterial line was placed for blood pressure monitoring and sampling. He required a low-dose phenylephrine infusion at the maximum rate of 450 μg/hour to maintain the target mean arterial pressure.
Anaesthesia included target-controlled infusion (TCI) of remifentanil and 8 mg of intravenous oxycodone. He received a total of 2250 mL of compound sodium lactate; his urine output throughout the surgery was 60 mL/hour (1 mL/kg/hour), and estimated blood loss was 500 mL. Postoperative analgesia included a right paravertebral nerve block infusion at T6–T7, which was inserted prior to the surgery, and a patient-controlled analgesia (PCA) morphine.
Intraoperative arterial blood gas (ABG) showed high anion gap metabolic acidosis (HAGMA) (table 1). As the lactate levels were normal, a formal renal panel and plasma ketones were evaluated. He was found to have high ketones. The beta-hydroxybutyrate (BOHB) was 5.5 mmol/L, with normal glucose. The patient was extubated after a short spontaneous breathing trial and was haemodynamically stable in the postoperative care unit. He showed no signs of dehydration. Appropriate respiratory compensation was observed as the patient was alert and pain free. Postoperatively he was admitted to the high dependency unit with an intravenous insulin infusion with a 5% dextrose drip of 100 mL/hour required for the next 36 hours before his ketones and acidosis normalised (table 1). His pain management, sleep and oral intake remained optimal over the subsequent days until discharge.
Summary of arterial blood gas results and beta-hydroxybutyrate levels
| Arterial blood gas analysis | Intraoperative | Post Anaesthesia Care Unit (PACU) | 6-Hour postoperative | 24-Hour postoperative |
| pH | 7.218 | 7.294 | 7.313 | 7.411 |
| pCO2 (mm Hg) | 37.5 | 28.8 | 24.9 | 31.8 |
| pO2 (mm Hg) | 318.5 | 74.5 | 64 | 94.6 |
| Bicarbonate (mmol/L) | 14.9 | 13.7 | 12.3 | 19.7 |
| Base excess (deficit) | −12 | −11.4 | −12.2 | -4 |
| Haemoglobin (mg/dL) | 12.9 | 12.4 | 12 | 11.2 |
| Oxygen saturation (%) | 99 | 93.9 | 91.6 | 96.8 |
| Sodium (mmol/L) | 138 | 137 | 137 | 134 |
| Potassium (mmol/L) | 4.5 | 4.3 | 3.9 | 3.8 |
| Chloride (mmol/L) | 110 | 109 | 111 | 110 |
| Anion gap | 17.6 | 18.6 | 17.6 | 8.1 |
| Lactate (mmol/L) | 1.7 | 2 | 1.9 | 1.3 |
| Glucose (mmol/L) | 7.8 | 9.1 | 9.8 | 11.5 |
| Beta-hydroxybutyrate (mmol/L) | 5.5 | 2.6 | ||
-
PACU, Post Anaesthesia Care Unit; pCO2, partial pressure of carbon dioxide; pO2, partial pressure of oxygen.
The patient was transferred to the general ward and discharged on postoperative day 6. When reviewed in the clinic 3 weeks later, the patient reported a weight loss of 15 kg. His fracture and wound healing were declared satisfactory. The patient was referred to the nutritionist for further evaluation and management.
Investigations
The patient’s ABG and ketone assays are summarised in table 1.
Differential diagnosis
This patient has no history of diabetes mellitus and is non-alcoholic. He was admitted for polytrauma with several rib, scapular and clavicle fractures and underwent definitive surgical fixation of his multiple fractures and split skin graft for degloving injury. Regular oral analgesics were administered during this period, and the reported pain scores were mild (0–3 on the numeric rating scale).
His oral intake has been reasonable, with nurses documenting three-fourths to one share of meals consumed at each serving during the time leading to the major surgery on the fifth day of admission to the hospital. He was not clinically acidotic, nor did he complain of anxiety.
His intraoperative ABG showed HAGMA with normal lactate and glucose despite normal bicarbonate and anion gap in the preoperative bloods. Subsequent blood tests done showed ketosis. At the time, it was attributed to starvation ketosis, although the fasting time was not extensive.
Since the patient did not have diabetes mellitus and was normoglycaemic, this clinical picture pointed toward extensive catabolism.
Euglycaemic ketoacidosis may occur in patients taking sodium–glucose cotransporter 2 inhibitors (SGLT2i) or patients with prolonged fasting, especially with bowel pathologies or severe pancreatitis. Our patient did not exhibit evidence of such conditions.
Treatment
As per treatment of ketoacidosis protocol, the patient was started on an insulin–dextrose regimen in view of euglycaemia until the ketoacidosis resolved. The acid–base balance on ABG, serum electrolytes, dextrose and ketone levels were monitored during the insulin infusion.
Outcome and follow-up
The patient remained well and participated in physiotherapy, mobilisation, rehabilitation and the patient-controlled analgesia (PCA) programme. His food intake has been adequate. He was eventually discharged and scheduled for regular follow-up with the orthopaedic, general surgery and plastic surgery departments.
Discussion
Stress response to trauma and surgery consists of the neuroendocrine and inflammatory responses.1 While every patient mounts a stress response to surgery, trauma or systemic illness, this could be exaggerated in cases of major trauma or surgery. The neuroendocrine response mainly relates to the regulation of the hypothalamic–pituitary–adrenal (HPA) axis, cortisol, and sympathetic stimulation give rise to an exaggerated response.
The paraventricular nucleus in the hypothalamus receives multiple central inputs, including the limbic area, which processes stressors; brainstem nuclei which receive physiological and inflammatory input; and the suprachiasmatic nucleus, which gives the HPA axis its circadian rhythm. The hypothalamus releases corticotropin-releasing hormone (CRH), which acts on the anterior pituitary gland to release adrenocorticotrophic hormone (ACTH), which stimulates the zona fasciculata in the adrenal cortex to synthesise de novo cortisol. The hypothalamus also stimulates the sympathetic nervous system and causes the release of epinephrine and norepinephrine from the adrenal medulla, which induces lipolysis, glycogenolysis and cellular metabolism.2
Cortisol exerts negative feedback on the hypothalamus and anterior pituitary to inhibit CRH and ACTH release, respectively. Cortisol is released in an ultradian pulsatile manner (high-amplitude pulses due to negative feedback by cortisol) with an overlying circadian rhythm (low night-time nadir, which progressively rises in anticipation of awakening).3 Cortisol dominates the body’s catabolic response to metabolic stress.
Tissue injury releases proinflammatory cytokines, including interleukin 1, interleukin 6 (IL-6) and tumour necrosis factor-alpha (TNF-alpha), leading to an inflammatory response. These cytokines lead to adrenal sensitisation and act on the hypothalamus to remove inhibitory tone on the pituitary gland, increasing the release of pituitary hormones such as growth hormone, vasopressin and ACTH. This ultimately leads to a rise in cortisol, which is further amplified as large cortisol pulses significantly increasing free cortisol levels due to the saturable nature of carrier protein.4–6 In patients experiencing perioperative stress, ACTH and cortisol levels will rise postoperatively, and ACTH levels will normalise at 24 hours; however, cortisol levels may remain elevated for 7 days, suggesting an uncoupling of the HPA axis.4 Cortisol shifts the body towards a hypermetabolic and catabolic state, promoting gluconeogenesis, muscle protein catabolism, lipolysis and anti-insulin effect.7 8
Insulin levels fall after general anaesthesia induction secondary to alpha-adrenergic inhibition of pancreatic beta cell secretion, while glucagon levels have a transient rise, further promoting a catabolic state.4 Increased catabolism, together with insulin deficiency, leads to ketogenesis.
An exaggerated stress response could result in severe HAGMA, weight loss, hyperglycaemic, surgical site infections, dehydration, increased length of hospital stay and poor outcomes.
Anaesthesia and the metabolic stress response
Various anaesthesia techniques can be used to mitigate the stress response. Techniques include the choice of anaesthetic drug and volatile anaesthetic, use of alpha-2 agonists such as clonidine and dexmedetomidine, opioids and regional anaesthesia. Surgical techniques include minimising tissue manipulation and injury by minimally invasive surgery where possible. Enhanced recovery after surgery programmes that aim to minimise fasting time, encourage early enteral nutrition and mobilisation postoperatively can also counter muscle protein catabolism from stress response.1
Anaesthetic drugs can have profound effects on the stress response. Etomidate causes adrenal suppression through reversible inhibition of 11-beta hydroxylase and 17-alpha hydroxylase, which inhibits cortisol synthesis up to 8 hours after a bolus dose.9 Despite reduced cardiovascular instability compared with other induction agents, the use of etomidate has become unpopular due to poorer outcomes in patients with sepsis.10
Propofol as total intravenous anaesthesia compared with volatile anaesthetics was able to reduce cortisol levels and glucose levels. When combined with remifentanil infusion, catecholamine levels were also reduced during the intraoperative and postoperative period.11–13 Compared with sevoflurane anaesthesia, the fat emulsion content of propofol may influence the body to metabolise the medium-chain triglycerides instead of protein, giving rise to protein-sparing effects.14
Volatile anaesthetics may have differences in their effect on the stress response; sevoflurane was associated with a reduction in catecholamine levels compared with desflurane. However, patients having desflurane anaesthesia had lower cortisol and ACTH levels.15 In a rat study, halothane, ketamine and thiopental were found to suppress the number and activity of natural killer cells, although propofol did not.16
Opioids are crucial in suppressing the HPA axis. Acute administration can increase CRH secretion and thus downstream ACTH and cortisol levels.17 However, opioids play a crucial role in analgesia hence diminishing the sympathetic drive in trauma patients and surgical patients. This seems to predominate, thereby diminishing the overall HPA activity. Intraoperative remifentanil infusion is able to blunt sympathetic response and reduce cortisol and epinephrine levels in the perioperative period.18 The ability of opioids to suppress stress and the sympathetic response has been especially beneficial for cardiac surgery, and it is often used in high doses. Morphine is able to block the secretion of growth hormone, while fentanyl, sufentanil and alfentanil reduce pituitary hormone secretion until cardiopulmonary bypass. In abdominal surgery, fentanyl can inhibit cortisol levels.4 Opioids have complex immunomodulatory effects, the mechanism of which is not fully understood. It is postulated that this could occur via effects on the HPA axis and the sympathetic system, or possibly via actions on both Mu and non-classical nociceptin/orphanin receptors on immunocytes. Morphine, fentanyl and remifentanil have a more substantial degree of immunomodulation than oxycodone and tramadol; this could indicate a central effect since opioids that cross the blood–brain barrier are stronger immunomodulators.17
Central alpha-2 agonists such as clonidine and dexmedetomidine bind to central presynaptic alpha-2 receptors, inhibiting sympathetic tone and outflow, thus decreasing circulating catecholamine levels. They have sedative, anxiolytic, analgesic and bradycardic effects. Despite its ability to reduce sympathetic outflow, it preserves the baroreceptor reflex, which is desirable in the perioperative setting.19
Dexmedetomidine has been shown to reduce catecholamines20 and cortisol rise to a similar extent as epidural analgesia.21 Dexmedetomidine attenuates the inflammatory process by reducing the perioperative rise in TNF-alpha and IL-6 levels and modulates T helper cell differentiation.19
It is more effective than labetalol and fentanyl at reducing the rise in heart rate and blood pressure during laryngoscopy and intubation. It reduces the sympathetic response to pneumoperitoneum during laparoscopic surgery to a greater extent when compared with fentanyl.22–24
However, these central alpha-2 agonists are associated with clinically significant bradycardia.25 Further studies are required to understand the role of dexmedetomidine on metabolic stress response and surgical outcomes.
Intravenous lignocaine infusion may attenuate the stress response by reducing postoperative levels of inflammatory mediators such as interleukin 6.26 It also has analgesic effects. However, cortisol levels remain unchanged, and further research is required on its effect on the stress response.27
Regional anaesthesia, especially epidurals, paravertebral blocks28 29 or transversus abdominis plane blocks30 in lower gastrointestinal surgeries, epidurals in spine surgery31 or orthopaedic surgeries,32 have been studied and show a reduction in catecholamine, cortisol and ACTH levels. In particular, epidurals have sympatholytic effects and block sympathetic chain. This mitigates the stress response via the HPA axis and locus coeruleus–sympathetic–adrenal medullary axis.
In a study where combined spinal-epidural (CSE) (with the epidural component providing postoperative pain relief) was compared with general anaesthesia with PCA postoperatively in patients undergoing hip arthroplasty, CSE led to an anticatabolic effect with reduced amino acid proteolysis, unchanged insulin levels and cortisol levels compared with the general anaesthesia group where there was a rise in amino acid oxidation, lower insulin levels and a rise in cortisol levels during and after surgery.33 Further research is required to determine which types of regional anaesthesia would reduce surgical stress response for which specific types of surgeries.
Benzodiazepines may inhibit steroid production at the hypothalamic–pituitary level, but the clinical significance of this action is uncertain.34
All trauma and surgical patients mount a metabolic stress response driving the body into a catabolic state with higher basal metabolic requirements. In rare cases, patients could mount a hypercatabolic response via neuroendocrine and inflammatory pathways promoting gluconeogenesis, lipolysis and ketogenesis, inverted glucagon: insulin ratio, leading to complications such as ketoacidosis. This hypermetabolic state is known to cause a reduction in muscle mass, wound infections and detrimental surgical outcomes.
Ketosis or hyperketonaemia is defined by BOHB levels exceeding 1 mmol/L. Further rise accompanied by low bicarbonate and high anion gap results in ketoacidosis. In most circumstances, starvation ketosis results in mild or nil acidosis. Extensive fasting or malabsorption could lead to severe ketoacidosis.35 Preoperative fasting for 6–8 hours is standard for most centres around the world, and the practice does not seem to increase the risk of fasting ketoacidosis. Losing 15 kg bodyweight perioperatively is also extremely uncommon in this patient group. Therefore, it is more prudent to conclude that this was caused by a hypermetabolic state.
A multitude of anaesthetic techniques can be used to attenuate the metabolic response to trauma. The use of central neuraxial regional anaesthesia is shown to suppress the HPA axis and sympathetic outflow. In addition, it provides excellent analgesia for postoperative recovery and rehabilitation. The use of potent opioids, remifentanil, in particular, has been shown to lower the sympathetic response. Propofol infusion, sevoflurane, desflurane and alfa-2 agonists have been shown to diminish the HPA activity and catecholamine levels.
We used a number of techniques discussed above in our patient’s anaesthetic care. These included multimodal analgesia with TCI remifentanil, intravenous oxycodone and regional anaesthesia in the form of a paravertebral nerve block catheter. Postoperatively the patient received PCA morphine in addition to the paravertebral nerve block infusion. The patient developed ketoacidosis secondary to a hypercatabolic state despite these efforts during the prolonged 10-hour surgery. However, it is unclear whether unique cytogenetic or pharmacogenetic factors played a role in this exaggerated stress response.
Prompt treatment with insulin and dextrose to correct the insulin:glucagon ratio and reverse the catabolic state of lipolysis, ketogenesis, protein catabolism, glycogenolysis, and re-establish glucose homeostasis.
We also feel that optimising nutrition with adequate calories and protein supplements together with minimised fasting may have reduced the magnitude of the stress response.
Learning points
-
Although the causation remains a matter of exclusion and thought provocation, this case report raises awareness of exaggerated stress response in patients with major trauma and/or surgery.
-
It is vital to anticipate and actively screen susceptible patients for ketoacidosis.
-
In recent years, many enhanced recovery programmes have been developed to promote anaesthetic techniques that attenuate the metabolic response to trauma/surgery and faster return to baseline physiological and functional state. Whenever possible, such techniques should be adapted to improve patient outcomes.
-
Optimisation of nutrition and minimising fasting time in patients with major trauma or surgery should not be neglected.
Ethics statements
Patient consent for publication
Footnotes
-
Contributors DSK conceived the idea of the case report and participated in the design, data collection, analysis, drafting of the manuscript, revision of the manuscript and approval of the final manuscript. YXC and NL participated in data collection, analysis, drafting of the manuscript, revision of the manuscript and approval of the final manuscript.
-
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
-
Case reports provide a valuable learning resource for the scientific community and can indicate areas of interest for future research. They should not be used in isolation to guide treatment choices or public health policy.
-
Competing interests None declared.
-
Provenance and peer review Not commissioned; externally peer reviewed.
- © BMJ Publishing Group Limited 2022. No commercial re-use. See rights and permissions. Published by BMJ.
References
Use of this content is subject to our disclaimer