Day-Four Fever With a Rising Haematocrit: Admit or Reassure?
A 26-year-old man in a Ranchi suburb attends a private clinic on the fourth day of fever. Temperature 39.2 °C, severe headache, pain behind the eyes, generalised body ache with calf tenderness. He has vomited once but is taking oral fluids. Blood pressure 104/70 mmHg, pulse 98/min, respiratory rate 18/min, no breathlessness. There is no gum bleeding, no black stools and no abdominal tenderness. A tourniquet test is positive. Reports: haemoglobin 15.2 g/dL, haematocrit 46%, platelets 88,000/µL, total white cells 3,400/µL; NS1 antigen positive, dengue IgM negative. Two years ago his haematocrit was 42%. It is the monsoon, and leptospirosis is also common locally. He lives 70 km away, and the only bus leaves at 6 pm; the family asks whether he can simply return tomorrow for another platelet count.
1. Which single laboratory finding shifts this from 'fever to be watched at home' to 'admit now', and why? Explain what is happening physiologically.
Read the worked answer
The decisive finding is the haematocrit: 46% against a documented baseline of 42%, a rise of 4 points, about 9.5%. Combined with platelets falling to 88,000 and white cells to 3,400 on day 4, this is plasma leakage into the interstitium - the start of the critical phase, which usually coincides with defervescence. Dehydration alone raises haematocrit but does not produce this triad, and he is drinking.
NS1 positivity confirms dengue; IgM is often still negative on day 4, so that does not exclude it. Leptospirosis is locally endemic and calf tenderness is compatible, but there is no conjunctival suffusion, jaundice or oliguria, and the NS1 result fits dengue.
The finding changes management because a falling haematocrit later, not a rising one, will mark the dangerous phase.
2. The family asks for a platelet transfusion because the count is 'low'. How do you decide, and what do you tell them?
Read the worked answer
Platelet count alone is not a transfusion trigger. At 88,000/µL with no mucosal bleeding, no petechiae beyond a positive tourniquet test and stable blood pressure, the risks of transfusion (reactions, TRALI, volume load, wasted units) exceed the benefit. Transfused platelets are consumed rapidly in dengue and do not prevent leakage. Reserve transfusion for clinically significant bleeding, or for very low counts with a bleeding tendency, and even then give judiciously.
What matters is fluid balance and watching for warning signs. Explain this plainly: the number is expected to fall for another day or two and then recover; the danger signs are bleeding, persistent vomiting, severe abdominal pain, drowsiness and reduced urine output. Also avoid intramuscular injections, aspirin and ibuprofen, which worsen bleeding risk.
3. Should he be sent home with a return ticket for tomorrow? Justify using the numbers and the logistics.
Read the worked answer
No. The combination of a rising haematocrit, thrombocytopenia and day 4 of fever puts him in the window in which shock can develop within hours. He lives 70 km away, the last bus is at 6 pm, and the family's plan depends on a night journey they cannot make; the single vomit also hints at poor oral intake.
Admit for 24-48 hours of observation with haematocrit, urine output, blood pressure and warning-sign checks. Give oral fluids if tolerated; use intravenous crystalloid only for poor intake or a rising haematocrit with a narrowing pulse pressure. Avoid intramuscular injections, aspirin and ibuprofen.
If admission is refused, arrange a documented daily review and a 24-hour contact, but that is clearly the weaker option.