Postoperative anaemia and lactic acidosis following open prostatectomy

Intended learning outcomes

1. Recognise retroperitoneal haemorrhage to be a relevant and often missed cause for postoperative haemorrhage.

2. Explore the causes for lactic acidosis and its relationship to low haemoglobin levels.

3. Describe the indications and pitfalls for the use of ultrasound and computed tomography imaging in postoperative haemorrhage.

Introduction

A 70-year-old male patient with a complex surgical history, notably a prior right hepatic lobectomy for hepatocellular carcinoma, underwent a robot-assisted radical prostatectomy. His comorbidities included type 2 diabetes treated with metformin, iron-deficiency anaemia and hypertension. Intraoperatively, the surgical team encountered dense adhesions, necessitating an early conversion to an open approach. Despite this complication, the procedure was completed with an estimated blood loss of approximately 250 ml.

Report

Following transfer to the Post-Anaesthesia Care Unit, the patient remained stable from the cardiovascular and respiratory standpoint. However, an early, sharp rise in lactate levels to 6.6 mmol/L was noted, accompanied by the development of metabolic acidosis, episodes of nausea and vomiting, and a decrease in haemoglobin (Hb) from 99 g/L to 79 g/L. Lactate levels stabilised following fluid boluses and the transfusion of one unit of packed red blood cells (PRBCs). Nonetheless, his haemoglobin level continued to drop steadily over subsequent days, with a lowest level of 70 g/L, necessitating a total blood transfusion of five units of PRBCs and one unit of platelets. Haemodynamics remained stable and no further rise in lactate was noted. All other laboratory investigations remained within normal range. Repeated comprehensive physical examinations, including digital rectal examination and nasogastric tube aspirate, failed to identify the haemorrhage source. A bedside abdominal ultrasound was negative for an intra-abdominal or pelvic haematoma.

Question to be answered

What causes a continuous drop in haemoglobin in an otherwise stable patient and is there a connection with the initially present lactic acidosis?

Discussion

The clinical vignette presents a complex postoperative course following conversion to an open prostatectomy, characterised by substantial and ongoing, unexplained anaemia with an initial sudden metabolic lactic acidosis. Integrating the full clinical context, the most robust explanation for the sustained drop in haemoglobin and its connection with the initial lactic acidosis was an occult retroperitoneal haemorrhage, leading to transient hypovolaemia and regional hypoperfusion alongside postoperative fluid loss due to ongoing nausea and vomiting.

The patient’s requirement for a cumulative transfusion of five units of PRBCs over 3 days necessitated a systematic evaluation of causes for postoperative anaemia [1]. Whereas haemodilution from intensive fluid therapy could account for the initial limited drop in haemoglobin, it could not explain the magnitude or persistence of the anaemia observed. Similarly, although the patient had a pre-existing history of iron-deficiency anaemia (low baseline iron stores), making him less tolerant to blood loss, the acute, significant decline in haemoglobin pointed towards active blood loss rather than chronic factors, such as bone marrow suppression or nutritional deficiencies. Causes related to increased consumption, such as disseminated intravascular coagulation or haemolysis, were rendered less likely, given the patient’s overall haemodynamic stability and the report of otherwise unremarkable laboratory investigations.

Therefore, the sustained and substantial decline in haemoglobin, despite the absence of external or localised signs of bleeding, strongly implied an occult haemorrhage, specifically into the retroperitoneal space.

Retroperitoneal haemorrhage is a well-recognised, although often missed, complication of pelvic surgery. The inability to detect a source via physical examination is often attributed to the anatomical compliance of the retroperitoneal space and the diagnostic limitations of bedside assessment. The occult clinical manifestation of a retroperitoneal haematoma often masks its presence. Unlike bleeding into the peritoneal cavity, which rapidly causes acute peritonitis and severe, localised pain, haemorrhage contained within the retroperitoneal space produces less dramatic, and often non-specific, symptoms. Although abdominal pain remains the most common presenting complaint, it is not always present or can be masked by routine postoperative analgesia. Classic, specific signs of significant haemorrhage, such as flank bruising (Grey–Turner’s sign), are generally delayed and rarely visible in the acute setting. Similarly, abdominal tenderness may be minimal or absent. Furthermore, whereas a highly localised sign, for example femoral neuropathy (due to compression of the lumbar plexus by the expanding haematoma), may offer a clue, it can be easily overlooked in the general context of the acute postoperative phase. Physical examination should include flexing a patient’s hips by drawing them towards the abdomen, which can often trigger pain [2].

The failure to identify the source of bleeding in the retroperitoneal space with a bedside ultrasound examination is common and attributed to both the unique anatomy of the retroperitoneum and the technical limitations of non-invasive ultrasound imaging. Bedside focused assessment with ultrasound is often unreliable in detecting deep retroperitoneal haematoma and frequently yields false-negative results. Technical difficulties can severely limit visualisation. This may be due to signal attenuation from overlying structures including bowel gas and subcutaneous emphysema, resulting from the open surgical conversion. Crucially, the nature of slow, continuous bleeding means that the haemorrhage often rapidly forms clots. Clotted blood is less anechoic than simple free fluid and is consequently not seen as free fluid, making differentiation from the surrounding retroperitoneal soft tissue challenging. Furthermore, the sensitivity of ultrasound requires a substantial accumulation of fluid. The literature suggests that a minimum volume of approximately 600 ml is required for reliable detection; therefore, slow bleeds may be missed if this threshold is not met acutely [3].

For these combined clinical and technical reasons, the definitive localisation of a suspected retroperitoneal haemorrhage requires more sensitive cross-sectional imaging, with a contrast-enhanced computed tomography (CT) scan remaining the modality of choice.

The haemodynamic stability of the patient, maintaining normal blood pressures while experiencing a steady haemoglobin drop, is consistent with a slow, continuous haemorrhage into the large, compliant retroperitoneal space, allowing for volume compensation and masking the blood loss over several days, especially as intensified fluid management including blood transfusion is continued. Treatment is often conservative, avoiding further surgical intervention and potential injury. Management consists of a balanced approach to the administration of clotting products, as well as administration of tranexamic acid. Tranexamic acid acts through the inhibition of fibrinolysis with increased haemostasis and is most effective if administered early in patients who are bleeding [4]. Self-tamponading of the ongoing haemorrhage with the surrounding tissue is the awaited goal, especially in haemodynamically stable patients and if no active bleeding source is detected with CT imaging.

The early, sharp rise in lactate level accompanied by metabolic acidosis, which subsequently stabilised with fluid boluses and one unit of PRBCs, represents type A lactic acidosis caused by transient tissue hypoperfusion [5]. At this early postoperative stage, several factors contributed to the hypoperfusion, including ongoing fluid loss from repeated vomiting, intraoperative fluid loss not fully substituted with IV fluids and possibly the beginning of a haemorrhage.

But could the slow reduction in haemoglobin content have been the single cause leading to the rise in lactate level? It is known that oxygen delivery is very strongly influenced by the haemoglobin content of blood. Using the oxygen delivery equation as an example, and assuming that cardiac output remains unchanged at approximately 5 L/min and the arterial oxygen partial pressure is normal, the oxygen delivery of a 70 kg patient with a Hb of 70 g/L would be 7 ml/kg/min. As the approximate oxygen consumption of the human body at rest is known to be approximately 3.5 ml/kg/min, even with the lowest measured haemoglobin of approximately 70 g/L, oxygen delivery would still be double that amount in our patient scenario and does not qualify as an explanation for increased lactates alone [6, 7]. This equation is, however, a global measurement of oxygen delivery and does not take account of regional blood flow that may, for expanding retroperitoneal haematoma, be related to tissue compression and local ischaemia.

Another potential explanation for the initial lactic acidosis, although less likely, would be the patient’s long-term treatment with metformin. Metformin is known to inhibit the Krebs cycle, the biochemical pathway that normally converts pyruvate originating from glucose to acetyl-CoA. With the Krebs cycle inhibited, pyruvate is predominantly converted to lactate [8]. To reduce perioperative risk for the development of lactic acidosis, particularly when there is the potential for renal impairment and reduced clearance of metformin, the drug is typically stopped before surgery and throughout the initial postoperative course. However, as lactate clearance was achieved with fluid management, no kidney injury identified this reason as deemed unlikely.

Conclusion

In conclusion, the lactic acidosis presented here was the acute physiological response to hypovolaemia caused by multiple factors including haemorrhage, dehydration due to inadequate fluid management and fluid losses through postoperative vomiting. The unexplained, cumulative drop in haemoglobin over subsequent days was the definitive consequence of the ongoing retroperitoneal blood loss. All increases in lactate should be investigated. When accompanied by cardiovascular instability or a drop in haemoglobin levels, postoperative blood loss should be immediately ruled out. When a source of bleeding is not apparent, a contrast-enhanced CT scan is the imaging modality of choice to localise the source and guide interventional or surgical management.

Declarations and conflicts of interest

Artificial intelligence declaration

Artificial intelligence tools were used solely for language polishing and formatting. All clinical content, interpretation, references and final revisions were undertaken and verified independently by the author.