Haemophagocytic syndrome during pregnancy: a case report and review of literature

Introduction

HLH is poisonous to cells by granular activated pathway gene mutations that cause increased cytotoxic activity in the natural killer cells and cytotoxic T lymphocytes; this mechanism can damage and release a large number of cytokines. earlier and prompt diagnosis of HLH is vital for pregnant women. The cases of HLH in pregnancy were extremely rare and only a very small number of cases have ever been reported. From this case, we learned HLH can be likely ignored and not be diagnosed promptly in the pregnant patient. Given the high morbidity and mortality of this disease in recent years, an expert consensus or clinical guideline should be developed.

Case report

The patient is a female patient, 22 years old, who was admitted to Affiliated Hospital of Jining Medical University on October 7, 2021 for ‘pregnancy at 29 weeks, jaundice for 1 week, cough for 6 days, and fever for 3 days’. One week before admission, the patient’s urine was yellow, and she was urinating a small amount of 100–200 ml urine/time, ∼3 times/day, with no urgency, frequency of urination, pain on urination, and no diagnosis or treatment. Six days before admission, the patient developed a cough, accompanied by expectoration, and a dry and sore pharynx, the patient had a production of sputum making it not easy for the patient to cough, and the cough was accompanied by bilateral costal margin and upper abdominal pain. This pain was accompanied by fatigue, and the patient started drinking more water than normal but still had little urine production and no diagnosis and treatment. The patient developed a fever three days before admission, and the patient’s maximum body temperature was 39 °C. The patient’s body temperature eventually dropped to the normal range by itself. The day before admission, her family saw that she had yellow sclera, and at 20:00, she developed a fever again, with a temperature of 40 °C, accompanied by nausea and vomiting. At that time, the patient was vomiting stomach contents. She went to the local Beicheng Hospital, where she was given fluid rehydration, cooling, and anti-inflammatory treatment (the specific details are unknown), Her test results back showed increased myocardial enzymes, transaminase, and bilirubin, and the patient’s gallbladder wall was rough. It was then recommended that the patient goes to the superior hospital to see a doctor, and the patient arrived at our hospital and was evaluated by the emergency physician with a physical examination.

After a relevant examination after admission, the platelet counts progressively decreased. She had continuous antibiotic therapy. Hepatitis B and hepatitis C antigen tests were negative, and the liver function was abnormal with serum cholinesterase 2925.0 U/L, alanine aminotransferase 486.2 U/L, aspartate aminotransferase 334 U/L, total bilirubin 89 µmol/L, and combined bilirubin 67 µmol/L. Emergency colour ultrasound showed normal liver size and shape, parenchyma echo uniform; Gallbladder wall thickened and coarse. Acute cholecystitis was initially diagnosed and treated conservatively. Gallbladder puncture and drainage or surgical treatment were performed when necessary, and magnesium isoglycyrrhizinate injection, polyene phosphatidylcholine, and ulinastatin were added for liver protection treatment. and on the evening of the hospital admission, her temperature rose. She then developed nausea, vomiting, accompanied by chills and fever, and she was cold. She was given rehydration support, but her symptoms were not alleviated after symptomatic treatment. She had a progressive decline in her blood pressure, and lethargy, she had a shock index >1, her venous filling was poor, and the doctors considered the presence of septic shock. After the contraindications were eliminated, a caesarean section of the lower segment of the uterus (scar uterus) + modified uterine harness operation was performed on October 8 under static inhalation combined anaesthesia, and the operation went smoothly. The patient’s treatment with liver protection drugs was continued, and her antibiotics were changed to cefoperazone sodium/sulbactam sodium as an i.v. drip treatment. MRI showed localised fatty liver or liver damage, splenomegaly, tortuosity and dilatation of splenic veins, edoema of Greenson’s sheath, and cholecystitis. Glutathione was added to protect liver function. The parturient developed a recurrent fever on October 16th, and a scattered skin rash was seen on the patient’s chin and chest. The parturient was treated with imipenem and statin sodium for infections again, but the treatment effect was poor. Blood imaging showed low white blood cell and neutrophil counts, and the patient’s ferritin level was 2312.85 ng/ml (October 13, 2021). Then, the patient was admitted again into the intensive medicine department on October 17, her antibiotics were changed to vancomycin and piperacillin, the patient’s body temperature was still high, and the patient had routine blood tests in three series, including a ferritin level of 1475.56 ng/ml. After full discussion, infection was considered to be the most likely aetiology, but a blood syndrome could not be ruled out. On October 22, Imide was administered based on the culture results, and the infection was resistant to vancomycin. The patient was given prednisolone to control inflammation, glutathione, and magnesium isoglycyrrhizinate were injected to protect the liver. The patient received plasma exchange, transfusions, and rehydration therapy on October 21. The vancomycin was discontinued, and the patient was given ring element 100 mg as a static drop q12 h to enhance her anti-infection treatment. Bone marrow biopsy showed hyperplasia of two granulated cell lines, and abnormal lymphocytes could be seen, with haemophagocytosis. The patient was then confirmed to have haemophagocytic syndrome. Rituximab 100 mg was administered twice on October 21 and 28, and the patient’s body temperature then became stable. PET/CT examination was performed, and the results were that the bone shape of the axial bone was fair, and the metabolism was increased. The patient and her family refused to undergo genetic testing. After continued treatment, the patient’s condition gradually improved, the antibiotics were gradually discontinued, and the methylprednisolone was gradually reduced. A routine blood examination was performed on October 29, 2021, and a colour ultrasound and other auxiliary examinations were performed. The patient’s condition improved, and she was discharged.

The final diagnosis at discharge was the following: haemophagocytic syndrome, sepsis, cholecystitis, a foetal head position of G4P12 at 29 weeks of gestation, bronchopneumonia, scarred uterus, obesity, pelvic inflammatory disease, pelvic adhesion, hypoproteinemia, extremely premature infant, neonatal respiratory distress syndrome, and an extremely low birth weight infant.

Literature review and discussion

Haemophagocytic lymphohistiocytosis (HLH) is a disease that is difficult to diagnose and treat (Ramos-Casals et al. 2014). HLH can be divided into primary (familial) and acquired (secondary) types (Janka et al. 1998). Primary HLH mostly occurs in children. It is inherited and characterised by mutations in the genes that encode proteins that are part of the signalling system involved in the inactivation of the immune system. Acquired HLH is the main cause of adult HLH, and various infections, such as viruses and bacteria, autoimmune diseases, and malignant tumours, are the triggering factors (Malinowska et al. 2014, Ramos-Casals et al. 2014).

The most widely accepted diagnostic model comes from the inclusion criteria of a groundbreaking clinical trial known as HLH 2004 (Henter et al. 2007). Only children were recruited for this trial, most of whom had the primary disease (Bergsten et al. 2017). The lack of rigorous data on adult secondary HLH results from the reliance on paediatric data for the diagnosis of adult secondary HLH. Therefore, these criteria also serve as the diagnostic basis for secondary HLH in adults. However, primary HLH (mostly seen in children) and secondary HLH (mostly seen in adults) are different diseases, and it is not correct to classify these diseases as a single entity and to adopt the same diagnostic criteria for them (Brisse et al. 2016).

The HScore is a risk-stratified diagnostic model, which is different from the pure dichotomy, such as HLH-2004, and it is a continuous HLH risk score, which can be converted into the probability of HLH (Table 1) (Fardet et al. 2014).

Table 1. Scoring systems for the diagnosis of HLH.

Symptom	HLH-2004	HScore
Fever	0 (<38.5) or 1 (≥38.5)	0 (<38.4), 33 (38.4–39.4), or 49 (>39.4)
Hepatosplenomegaly	0 (No) or 1 (Yes)	0 (none), 23 (hepatomegaly), or 38 (hepatomegaly)
Cytopenia	0 (one series) or 1 (two or three departments)	0 (one department), 24 (two departments), or 34 (three departments)
Ferritin	0 (<500) or 1 (≥500)	0 (<2000), 35 (2000–6000), or 50 (>6000)
Triglycerides	0 (<3) or 1 (≥3)	0 (<1.5), 44 (1.5–4), or 64 (>4)
Fibrinogen	0 (>1.5) or 1 (≤1.5)	0 (>2.5) or 30 (≤2.5)
Bone marrow blood phagocytosis	0 (No) or 1 (Yes)	0 (No) or 35 (Yes)
Aspartate aminotransferase	–	0 (<30) or 19 (≥30)
Known potential immunosuppression	–	0 (No) or 18 (Yes)
Scores^a	≥4 (100%)^b	90 (>1%)^c 100 (1%) 110 (3%) 120 (5%)
		130 (9%) 140 (16%) 150 (25%) 160 (40%)
		170 (54%) 180 (70%) 190 (80%) 200 (88%)
		210 (93%) 220 (96%) 230(98%) 240 (99%)
		250 (>99%)

HLH: haemophagocytic lymphohistiocytosis.

^aThe sum of the above nine symptoms scores.

^b100% indicates a definite diagnosis.

^cThe probability of disease occurrence corresponding to the score.

As seen in Table 1, seven of the nine criteria are the same as the HLH-2004 criteria. In the other two additional criteria, the first one is ‘known potential immunosuppression’, which has a relatively low impact on the total risk score (because it is the lowest risk score among the nine components). The second is an AST > 30 U/L, which is non-specific because this criterion may occur in almost any hospitalised adult patient (Naymagon 2021). However, a retrospective study showed that the threshold level of ferritin used in the diagnostic criteria for adult secondary HLH was too low and should be increased. This suggests that the other criteria used to diagnose HLH in adults should be re-evaluated (Naymagon et al. 2020).

Despite the shortcomings of the HLH-2004 criteria and HScore, retrospective case–control studies have shown that both of the diagnostic models are highly discriminatory in the identification of haemophagocytic syndrome in critically ill patients (Valade et al. 2021). In the same study, the diagnostic efficacy of the HLH-2004 guidelines was compared with the new HScore, with a sensitivity and specificity of 100 and 80% for children and 90 and 79% for adults, respectively (Debaugnies et al. 2016). Therefore, the HScore is more effective than the HLH-2004 guidelines in identifying HLH in both children and adults, but there is little difference in the specificity of the diagnostic utility between the two methods.

If patients have a complicated disease, multiple symptoms, and atypical symptoms, this disease can be difficult to diagnose. According to the HLH-2004 diagnostic criteria, our patient presented with a high fever as the main symptom, and the PET/CT examination suggested a large spleen, progressive decrease in the peripheral blood cells, a progressive decrease in the platelets, and a ferritin level >500 mg/L. A bone marrow puncture showed haemophagocytic phenomenon. Compared with the diagnostic criteria of 2004, six of the criteria for haemophagocytic syndrome were met. Based on the HScore, the patient had a baseline score of 194, which corresponded with an 80% chance of developing the disease. Some studies have found that the best predictive accuracy of an HLH diagnosis is the presence of four critical values (95.0% sensitivity and 93.6% specificity) of the HLH 2004 standard and if the HScore is >168 points (100% sensitivity and 94.1% specificity) (Knaak et al. 2020). Therefore, in this case, the patient was diagnosed with haemophagocytic syndrome.

It is worth noting that this patient’s diagnosis at discharge was sepsis, which is a life-threatening organ dysfunction caused by the dysregulation of the human body’s response to infection (Singer et al. 2016). However, the clinical manifestations of sepsis are non-specific, and currently, there is no gold standard to define the disease. Therefore, its clinical diagnosis and severity assessment are very difficult. Both HLH and sepsis can cause fever, leukopoenia, an elevated ferritin level, a low fibrinogen level, and thrombocytopenia caused by DIC, and these can especially cause worsening clinical conditions (Machowicz et al. 2017). The two diseases often occur simultaneously, and it is also believed that infection-related HLH may be the special manifestation or terminal stage of SIRS or may be due to the sepsis inflammatory response. A clinicopathological analysis was conducted on 70 autopsy cases in Fukui University Hospital, and haemophilia was detected in 29 cases (41.4%) (Inai et al. 2014), most of which were complicated by blood diseases and sepsis, which causes great difficulty in the diagnosis of the disease. The treatment of sepsis is basically anti-infective treatment and intensive symptomatic treatment (Esposito et al. 2017), but this treatment does not include treating the key factors of HLH. While some treatments may be effective for both conditions, the main principles are different. Therefore, an accurate diagnosis of these critically ill patients is crucial because then the best treatment can be given to save the patients’ lives.

Haemophagocytic syndrome can be distinguished from sepsis in the following ways. First, in terms of the pathogenesis and pathology, HLH is poisonous to cells by granular activated pathway gene mutations that cause increased cytotoxic activity in the natural killer cells and cytotoxic T lymphocytes; this mechanism can damage and release a large number of cytokines; and sepsis is a pathogenic disease that stimulates the immune system, which amplifies cloned antigen-specific T lymphocytes, inflammatory cytokines, activated cytotoxic T lymphocytes and macrophages (Allen and Mcclain 2015). Both diseases are associated with the release of a large number of cytokines, but the types of cytokines released during the two diseases are different. Studies have shown that the serum levels of IL-10, IL-12, and INF-γ in the HLH group are significantly higher than those in the sepsis group, while the levels of IL-1β and IL-6 in the sepsis group are significantly higher than those in the HLH group (Huang 2018). At the same time, a domestic study regarding the cytokine diagnostic criteria for HLH showed that an IFN-γ level >100 ng/L and an IL-10 level > 60 Ng/L were important, and the level of IFN-γ was higher than that of IL-6. The sensitivity, specificity, positive predictive value, and negative predictive value for HLH were 88.0, 98.7, 93.6, and 97.5%, respectively (Xiaojun et al. 2011). Therefore, a differential diagnosis can be made by measuring the levels of each cytokine. Meanwhile, a foreign study showed that when IFNγ > 75 pg/mL and IL-10 > 60 pg/mL, the sensitivity and specificity of an HLH diagnosis were 98.9 and 93%, respectively (Griffin et al. 2020). Despite the differences in the data criteria and the specificity and sensitivity of the results, both studies demonstrate that determining cytokine levels are beneficial in distinguishing between haemphagocytic syndrome and sepsis. Second, Clinical manifestations: Hematophagy syndrome and sepsis both presents with certain clinical manifestations, such as splenomegaly, and studies have shown that 75% of sepsis patients who have received antibiotic treatment do not have splenomegaly, and most of these patients have received antibiotic treatment before the definitive diagnosis. Therefore, to some extent, we can differentiate haemophagocytic syndrome from sepsis by looking for the presence of splenomegaly after antibiotic treatment (Arismendi-Morillo et al. 2004). Third, Laboratory tests: Fibrinogen can be used as a laboratory differentiator. Studies have shown that hypofibrinogenemia is present in approximately two-thirds of HLH patients, while the other clotting factors are usually within the normal range and are not observed to be elevated in sepsis patients (Janka 2012). Therefore, the measurement of fibrinogen can be used to distinguish sepsis from haemophagocytic syndrome (Madkaikar et al. 2016).

Conclusion

The above methods can only distinguish the two diseases to a certain extent through the observation and comparison of the clinical manifestations and laboratory examinations, and there is no gold standard diagnostic test for the differential diagnosis. In addition, current studies cannot make a unified standard for differential diagnosis, and further clinical research and observations are still needed.

Ethical approval

Ethics registration was deemed unnecessary due to the nature of the study.

Consent form

Written informed consent for publication of their details was obtained from the patient/study participant/parent/guardian/next of kin.

Author contributions

Xinying Liu reviewed the literature and contributed to manuscript drafting with Hui Yu, Gege Li, and Tiantian Yu. Xueqin Feng and Dongmei Man revised the article and had primary responsibility for the final content. All authors read and approved the final manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Nature and Science Foundation of China (82101797), the Natural Science Foundation of Shandong Province (ZR2021QH131), and PhD Research Foundation of Affiliated Hospital of Jining Medical University (2021-BS-025).