From the cumulative blood loss threshold to 4T etiology, from active third-stage management to the three-hour rule of tranexamic acid and massive transfusion targets; Guideline-based management of the most common lethal cause of obstetric emergencies.
⚕️ Target audience: This clinical review is intended for physicians and healthcare professionals; It is a summary of current literature and guidelines for educational purposes. Clinical decisions should be made based on patient-specific evaluation and applicable institutional protocols.
Postpartum hemorrhage (PPH) is an obstetric emergency and is among the top five causes of maternal death in both resource-rich and resource-limited countries. It is not advanced technology that determines the result, but timely recognition, a prepared team and a protocol that escalates the steps without delay. In California Pregnancy-Associated Mortality Review 2014–2018 data, 63% of deaths due to obstetric hemorrhage were found to be preventable, and a contributing factor was identified in 89% of cases: delayed response or escalation 78%, delayed recognition 72%, inadequate blood product use 53% (California Pregnancy-Associated Mortality Review, 2014–2018).
Primary (early) PPH occurs within the first 24 hours, and secondary (late) PPH occurs between 24 hours and 12 weeks. The classic definition—estimated blood loss ≥500 mL in vaginal delivery, ≥1000 mL in cesarean section—is problematic: bleeding may not be externally visible (intraabdominal, retroperitoneal, or in the pelvic floor), blood in the collecting vessels mixes with amniotic fluid, and postpartum morbidity is relatively rare in the 500–999 mL range. ACOG therefore revised its definition in 2017 (ACOG Practice Bulletin 183, 2017): cumulative blood loss ≥1000 mL or bleeding with signs and symptoms of hypovolemia within the first 24 hours. The criterion applies to both routes of birth; the same document emphasizes that a loss of >500 mL in vaginal delivery—especially if bleeding continues—should be considered abnormal (ACOG PB 183, 2017).
Here is where institutions differ: ACOG (PB 183, 2017) uses a cumulative criterion of ≥1000 mL or evidence of hypovolemia. RCOG Green-top Guideline No. 52 (2nd ed., 2016; BJOG 2017) retains the classic definition: “genital blood loss of 500 mL or more within 24 hours after birth”; The 3rd edition is in preparation. UpToDate does not consider the numerical threshold alone sufficient and bases the diagnosis on the combination of more than expected bleeding + findings of hypovolemia — when bleeding is occult, findings of hypovolemia may occur before visible blood loss (UpToDate 2026).
“Expected” blood loss can be inferred from prophylaxis studies: in vaginal deliveries receiving routine prophylaxis, less than 10% of patients will lose ≥500 mL, less than 2% will lose ≥1000 mL; in cesarean section, >500 mL in 63% and >1000 mL in 30% (UpToDate 2026). There is also no standard for severity classification—in a review of more than 300 studies from 1960–2024, severe PPH was generally defined as a loss of ≥1000 mL rather than ≥500 mL. Better classifications look at symptoms, route of delivery, whether bleeding is ongoing, and gradual thresholds, rather than a single volume; The California Maternal Quality Care Collaborative (CMQCC) scheme follows this logic (CMQCC Toolkit, 2022).
| Stage | Criteria | Intervention |
|---|---|---|
| Stage 0 | Every patient in labor/birth | Prophylactic oxytocin, quantitative measurement of cumulative blood loss, close monitoring |
| Stage 1 | Cumulative loss ≥500 mL at vaginal delivery or ≥1000 mL at cesarean section; bleeding continues or there is evidence of occult bleeding (pulse ≥110/min, TA ≤85/45 mmHg, O 2 saturation <95%, confusion or shock index >0.9) | Activate enterprise PPH protocol |
| Stage 2 | Bleeding or vital sign instability persists, cumulative loss <1500 mL | Next step in the protocol; mobilize rapid response team and blood bank support |
| Stage 3 | Bleeding continues; cumulative loss >1500 mL or >2 U red blood cell transfusion or abnormal vital signs or suspicion of DIC | Initiate massive transfusion protocol, switch to surgical approaches for bleeding control |
The ATLS classification (class I–IV) is derived from the non-pregnant population; nevertheless, it is reminiscent of clinically critical class II bleeding (15–30% of blood volume): tachycardia (100–120/min) and tachypnea (20–24/min) occur while the systolic pressure remains almost unchanged. The combination of increased pulse and tachypnea with stable systolic pressure should be considered compensated shock and the PPH protocol should be initiated even if only mild vaginal bleeding occurs (ATLS; UpToDate 2026).
PPH is reported in 1–3% of births; however, most of these figures are based on subjective estimation—prospective studies using quantitative measurements have reported a loss of ≥500 mL in up to 10% of births. In the US National Inpatient Sample data, the rate increased from 2.7% in 2009 to 4.3% in 2019; The increase in severe PPH is at least partially attributed to increased cesarean section rates and the associated risk of placenta previa and placenta accreta spectrum (PAS) (US National Inpatient Sample 2009–2019; UpToDate 2026).
After separation of the placenta, hemostasis is achieved by two mechanisms: mechanical hemostasis (myometrial contraction compresses the vessels of the placental bed) and local thrombosis. The disorder has a high potential to develop into massive bleeding: in late pregnancy, uterine artery blood flow is 500–700 mL/min (60 mL/min in the non-pregnant) and approximately 15% of cardiac output (UpToDate 2026).
PPH causes are classified into the 4Ts: tone (uterine atony), trauma (laceration, rupture, incision), tissue (retained tissue, clot, PAS), and thrombin (coagulopathy); approximately 8% of patients have more than one cause. A systematic review covering 327 studies and more than 800 million patients reports the distribution separately for all cases and severe/refractory cases (systematic review; UpToDate 2026).
| Cause (4T) | All PPH cases | Severe/refractory cases | Note |
|---|---|---|---|
| Tone — uterine atony | 70.6% | 41.4% | Approximately 1 in 40 births in the United States; indication for 27% of peripartum hysterectomies. Frequently responds to uterotonics |
| Trauma — laceration, rupture | 16.9% | 12.8% | It is the most common cause (55%) in massive PPH series (>2500 mL and/or ≥5 U erythrocytes; 349 cases); atonia alone 23% |
| Tissue — retained placenta | 16.9% | 13.8% | Diagnosis is made if the placenta does not come out 30 minutes after birth |
| Tissue — abnormal placentation (PAS, previa) | 3.9% | 8.8% | Its share doubles in severe cases; usually recognized by prenatal ultrasound |
| Thrombin — coagulopathy | 2.7% | 1.1% | Approximately 1 in 500 births in the United States; Late findings due to consumption and hemodilution in patients without bleeding diathesis |
The table carries two caveats. First, frequent is not the same as severe: the share of trauma and abnormal placentation increases with severity. Second, coagulopathy is a late finding—in a study of more than 18,000 patients with ≥1500 mL loss at birth, the median prothrombin time was 12.3 seconds, the longest aPTT was 30.4 seconds, and the lowest fibrinogen was 360 mg/dL; All are within the normal range for pregnancy. Thresholds for suspicion of coagulopathy in acute PPH: fibrinogen <300 mg/dL, platelets <100,000/µL, INR >1.5 and/or prolonged aPTT (UpToDate 2026).
The most established risk factors for atony are previous PPH and prolonged labor; Chorioamnionitis, therapeutic magnesium sulfate, induction/augmentation, uterine hyperdistention due to myoma and multiple pregnancy/macrosomia/polyhydramnios are added. Focal atony is a particular pitfall: the lower uterine segment may be dilated and atonic while the fundus contracts well; It is not noticed during abdominal examination, but is detected during vaginal examination or cesarean section. Since it can be intermittent or segmental, intervals in which bleeding subsides may mislead the clinician into thinking "massage is working" and delay definitive treatment (UpToDate 2026).
The same review separates risk factors by strength of association:
Risk scoring does not predict PPH — it organizes preparedness. In a review evaluating 14 prognostic models, none had been adequately validated in the general obstetric population. In a study comparing the CMQCC, ACOG Safe Motherhood Initiative, and AWHONN tools in more than 11,000 patients, fewer than 5% of high-risk patients developed significant PPH (3.9–4.6% versus 1.5–1.8% in the low-risk group). In CMQCC v1.0 validation, severe PPH occurred in 0.8%, 2.0%, and 7.3% of the low-, medium-, and high-risk groups, respectively; however, only 22% of severe PPH cases came from the high-risk group, and in one study 40% of patients who developed PPH had no risk factors at all. Even so, using a risk tool is better than not using one; in the United States, The Joint Commission mandates the use of an evidence-based tool upon admission to both the labor unit and the postpartum unit (CMQCC 2022; The Joint Commission).
Previous PPH alone is not a strong indication for screening for hereditary bleeding diathesis. However, unexplained PPH that does not respond to general measures—especially if there is a history of heavy menstrual bleeding, excessive bleeding after minor trauma, or a family history of bleeding disorders—should suggest a diathesis; These patients are particularly at risk because von Willebrand factor levels rise during pregnancy and decline rapidly after birth (UpToDate 2026).
Active third stage management is a bundle of interventions: prophylactic uterotonic (± tranexamic acid) immediately after the anterior shoulder or baby is born, followed by gentle controlled cord traction until the placenta is separated and expelled. The most important component is pharmacotherapy; The contribution of cord traction is modest. In the Cochrane meta-analysis, active management compared to expectant (physiological) management reduced: PPH >1000 mL 8 versus 24 per thousand (RR 0.34; 95% CI 0.14–0.87), PPH >500 mL 50 versus 145 per thousand (RR 0.34; 0.27–0.44), postpartum hemoglobin <9 g/dL 36 per thousand versus 71 per thousand (RR 0.50; 0.30–0.83), maternal transfusion 1% versus 2.9% (RR 0.35; 0.22–0.55), therapeutic uterotonic use 3.9% versus 21.1% (RR 0.19; 0.15–0.23), and mean blood loss −79 mL (−96 versus −62) — Begley 2019, Cochrane.
The benefit is most pronounced in high-risk patients; The quality of evidence is low and benefit is uncertain at low risk. There are also adverse effects: increased return to hospital due to manual placenta removal, vomiting, after-pain, and post-discharge bleeding—each in 2–7% of patients, RR 2.2–3.6. The evidence for uterine massage is weak: in a meta-analysis combining three randomized trials (3842 singleton pregnancies) in patients who had already received oxytocin prophylaxis and controlled cord traction, massage did not reduce PPH (≥500 mL: 5.9% vs 4.0%; RR 1.09, 0.33–3.64; low-quality evidence). There are no randomized data after cesarean section; Because no harm has been shown, many clinicians continue massaging until the uterus hardens and remains firm (UpToDate 2026).
In a Cochrane network meta-analysis of 122 randomized trials and more than 120,000 participants, uterotonics were all found to be superior to placebo (Gallos 2025, Cochrane network meta-analysis).
| Medicine | Dose and route | Efficacy (PPH ≥500 mL) | Contraindication / caution |
|---|---|---|---|
| Oxytocin | IV infusion preferred: 10–40 U / 500–1000 mL 0.9% NaCl, up to 500 mL/hour. CMQCC: 10–30 U/500 mL or 20–60 U/1000 mL (typical: 30 U/500 mL, 10 U in the first 30 minutes, then 125 mL/h maintenance). If there is no IV line, 10 U IM | RR versus placebo 0.58 (0.48–0.69). IV superior to IM: 5.6% vs 7.2% (RR 0.78, 0.66–0.92); transfusion 0.6% vs 1.3% (RR 0.44, 0.26–0.77) — Oladapo 2020, Cochrane | The only contraindication is hypersensitivity. IV bolus is unnecessary for prophylaxis—rapid injection can lead to hypotension, myocardial ischemia, and cardiovascular collapse (CNGOF recommends 5–10 U over at least 1 minute if a bolus is to be given, and at least 5 minutes if there is cardiovascular risk). Long-term high doses may cause hyponatremia; May prolong QT |
| Methylergonovine (methylergometrine) and ergot alkaloids | 0.2 mg IM or directly into the myometrium; Onset of effect 2–5 minutes. Oxytocin–ergometrine combination (5 U oxytocin + 0.5 mg ergometrine, IM) is not available in the USA | Ergometrine alone RR 0.67 (0.50–0.91); trend towards lower efficacy compared to oxytocin (RR 1.16, 0.89–1.53). Ergometrine + oxytocin RR 0.44 (0.35–0.54) | 🔴 Contraindicated in hypertension, Raynaud's phenomenon, known coronary artery disease and hypersensitivity. Slow IV injection is not recommended due to the risk of serious cardiovascular/cerebrovascular events. Increased blood pressure, headache and vomiting are common |
| Carbetocin | 100 mcg IV over 30–60 seconds; The same dose can be given IM and be more effective. Cumulative maximum 100 mcg; No subsequent infusion required due to long effect | RR versus placebo 0.57 (0.43–0.74); As effective as oxytocin (RR 0.98, 0.79–1.22) | Half-life 40 minutes (4–10 times that of natural oxytocin). Its side effect profile is similar to oxytocin (including QT prolongation). Not licensed in the USA |
| Misoprostol (PGE 1 ) | FIGO and WHO recommend 600 mcg oral or sublingual if injectable uterotonics are not available. Vaginal route is not practical postpartum (bleeding impairs absorption) | RR versus placebo 0.64 (0.53–0.77). Compared with oxytocin, results are mixed and worse than oxytocin for PPH ≥1000 mL (RR 1.24, 1.06–1.46) | It is heat resistant, does not require needles. Chills and fever are common: fever begins in 20 minutes, peaks in 1–2 hours, and subsides in 3 hours. With 200–400 mcg sublingual, 8% of patients had fever >39 °C, and with 600 mcg, 45% had fever. |
| Tranexamic acid | 1 g IV over 10 minutes. After cord clamping during vaginal birth; If used in cesarean section, before skin incision | In vaginal delivery, in a 2025 Cochrane meta-analysis (3 trials, ~19,000 participants), it was not superior to placebo for PPH ≥500 mL and ≥1000 mL, transfusion, or severe morbidity; it reduced the need for additional uterotonics in women without anemia (7.3% vs 9.7%; RR 0.75, 0.61–0.92) | Contraindicated in known thromboembolic events during pregnancy, history of coagulopathy, active intravascular coagulation and hypersensitivity. In a meta-analysis with individual patient data in pregnant women (5 studies, >54,000 patients), thromboembolic events were 0.2% in both groups (OR 0.96, 0.65–1.41). |
In the network meta-analysis, combinations were more effective than oxytocin: oxytocin + ergometrine with PPH ≥500 mL 8.7% versus 11.5% (RR 0.76; 0.64–0.90), oxytocin + misoprostol with 8.1% versus 11.5% (RR 0.70; 0.57–0.87); transfusion risk was reduced by 27% (RR 0.73; 0.56–0.96) and 60% (RR 0.40; 0.28–0.58), respectively (Gallos 2025, Cochrane). The cost is side effects: vomiting with oxytocin + ergometrine 6.3% versus 2.0% (RR 3.15; 2.24–4.43) and hypertension 11.0% versus 7.3% (RR 1.51; 0.82–2.78); fever with oxytocin + misoprostol 9.1% versus 2.3% (RR 3.95; 2.52–6.20). ACOG and SMFM have not recommended combination prophylaxis—in the US, the standard is oxytocin alone (ACOG; SMFM). UpToDate recommends adding a second agent (tranexamic acid, misoprostol, or methylergometrine) to oxytocin in the high-risk patient [Grade 2C]. WHO and FIGO recommend 600 mcg oral/sublingual misoprostol in settings where injectable uterotonics are not available, and heat-stable carbetocin if there is no cold chain (WHO; FIGO).
The place of thermostable carbetocin was identified by the CHAMPION noninferiority study: IM carbetocin 100 mcg immediately after vaginal birth compared with oxytocin 10 U, finding noninferiority with blood loss ≥500 mL or additional uterotonic use in 14.5% versus 14.4% (RR 1.01; 0.95–1.06). verified. Loss ≥1000 mL was similar (1.51% vs 1.45%; RR 1.04, 0.87–1.25), but noninferiority could not be demonstrated for this outcome because the upper limit of the confidence interval exceeded the margin—most likely due to insufficient power (CHAMPION, NEJM 2018).
A 2023 meta-analysis (50 studies) reported that prophylactic tranexamic acid reduced blood loss >1000 mL from 45.5% to 11.03% (RR 0.26; 0.17–0.42) in high-risk patients and from 13.8% to 11.1% (RR 0.64; 0.51–0.81) in low-risk patients; timing produced a significant subgroup difference (RR 0.33, 0.25–0.44 before skin incision; RR 0.86, 0.79–0.93 after cord clamping). In contrast, two large randomized trials did not show benefit in the low-risk population: in a trial involving 11,000 patients across 31 hospitals where 1 g was given after cord clamping, the composite of maternal death or transfusion was 3.6% versus 4.3% (adjusted RR 0.89; 0.74–1.07) and blood loss >1000 mL was 7.3% versus 8.0% (RR 0.91; 0.79–1.05); only the need for intervention for bleeding complications decreased (RR 0.90; 0.82–0.97 — Pacheco 2023, NEJM). A 2024 Cochrane review (6 trials, 15,981 participants) found a reduction in calculated blood loss ≥1000 mL (RR 0.83; 0.76–0.92) but no difference in estimated blood loss (RR 0.94; 0.79–1.11) or transfusion (RR 0.88; 0.72–1.08) (Rohwer 2024, Cochrane). A plausible explanation for the discrepancy is timing—the two large trials administered the drug after cord clamping. If prophylactic tranexamic acid is used during cesarean delivery, administration before skin incision may be argued for, although fetal/neonatal safety data for this timing are lacking (TRAAP2, NEJM 2021; UpToDate 2026).
Quantitative blood loss (QBL) is the systematic use of volumetric cups and scales or computerized image recognition; Estimated blood loss (EBL) is a visual assessment and is less sensitive and specific. Methods: volumetry (calibrated pocket drape, calibrated aspirator cup), gravimetry (wet–dry weight difference; gram ≈ milliliters), AI-assisted colorimetry (correlates well with validated reference but insufficient data for clinical recommendation) and visual companion cards (different from traditional visual EBL, requires regular training). In methods other than colorimetry, non-blood fluids (amniotic fluid, irrigation, urine) must be taken into account. In one systematic review, calibrated dressing + clinical observation (pulse, blood pressure, uterine tone) detected 93 of 100 patients with PPH and missed 7; It evaluated 5 out of 100 patients without PPH as false positive. However, objective QBL methods have not been proven to improve clinical outcomes (UpToDate 2026).
Institutional separation: AWHONN (Practice Brief 13, 2021) and UpToDate recommend QBL for all births. ACOG, however, did not make a recommendation favoring QBL over EBL. In its 2023 recommendations, WHO recommends objective measurement of blood loss and its combination with a standard treatment package to improve the detection and treatment of PPH in vaginal birth. Hemoglobin and hematocrit are poor indicators of acute blood loss: it may take up to four hours to see a laboratory change, rarely 48–72 hours (AWHONN Practice Brief 13, 2021; WHO 2023; ACOG).
Approximately 90% of deaths due to PPH occur within four hours of birth. The critical caveat is that blood pressure decline is a late finding: systolic pressure may not fall to <90 mmHg until the patient loses up to 25% of blood volume (≥1500 mL in pregnancy). Therefore, sustained bleeding of ≥500 mL in vaginal delivery and >1000 mL loss in cesarean section are alarm triggers; In the presence of abnormal vital signs, lower thresholds (e.g., ≥300 mL in vaginal birth) may provide earlier recognition (UpToDate 2026).
The combined effect of early detection and bundle therapy was demonstrated by the E-MOTIVE study, which included more than 210,000 vaginal births in 80 hospitals in four countries. Attempt; It consisted of early detection with a calibrated blood collection cover (trigger 300 mL in the presence of an abnormal clinical finding, 500 mL otherwise), first-line treatment bundle (uterine massage, uterotonic, tranexamic acid, IV fluid), examination of the genital tract and placenta, and escalation when necessary. Primary composite outcome (≥1000 mL blood loss, laparotomy for bleeding, or maternal death from bleeding) 1.6% versus 4.3% (RR 0.40; 0.32–0.50)—approximately 60% reduction. Two mechanisms producing the outcome were evident: PPH detection rate 93.1% versus 51.1% and compliance with the treatment bundle 91.2% versus 19.4% (E-MOTIVE, NEJM 2023).
The goals are to preserve circulating volume and tissue oxygenation, prevent or reverse coagulopathy, and treat the obstetric cause. What matters is not the exact sequence of drugs, but initiating uterotonic therapy without delay and rapidly evaluating its effect: within 30 minutes it should be clear whether pharmacologic therapy is reversing atony. If it does not, invasive interventions are initiated; hesitation leads to dilutional coagulopathy, hypovolemia, hypothermia, and acidosis, after which bleeding control becomes substantially more difficult (UpToDate 2026).
Call for help; if the patient is unstable or potentially unstable, transfer her rapidly to the operating room. Establish two intravenous lines, at least one large-bore (at least 18 G, preferably 14–16 G). Initiate resuscitation with isotonic crystalloid — one guideline recommends rapid infusion of up to twice the lost volume (up to 3.5 L) for blood loss >1000 mL or clinical shock; however, because >3–4 L of rapid crystalloid may cause dilutional coagulopathy, electrolyte abnormalities, and hypothermia, hematocrit, coagulation, core temperature, and electrolytes must be closely monitored.
To determine the cause, a full examination is performed in the lithotomy position, under adequate lighting and anesthesia: the entire vagina from the perineum to the cervix is examined for laceration (repeated even if it was examined during birth), uterine tone is evaluated, the cavity is checked for retained tissue and rupture, inversion and — especially in the patient who has attempted labor after cesarean section — uterine rupture are excluded. Fundal massage and bimanual uterine compression (one hand is made into a fist and placed vaginally through the anterior fornix, the other hand presses the fundus from the abdomen towards this hand) is continued while other interventions are prepared; If bleeding continues unabated while the fundus contracts well, the benefit of massage is unlikely (UpToDate 2026).
Tranexamic acid is given concurrently with other medications and procedures as soon as possible after the onset of bleeding. Dose: 1 g (10 mL of 100 mg/mL solution) IV over 10–20 minutes; Infusion faster than 1 mL/minute may cause hypotension. If bleeding continues after 30 minutes, the second 1 g can be applied. The antifibrinolytic effect lasts seven to eight hours in serum; The concentration in breast milk is approximately one percent of the serum peak. The evidence base is the WOMAN study (193 hospitals in 21 countries, more than 20,000 patients; 70% of births vaginal). Tranexamic acid compared to placebo:
The relevance of timing was even clearer in a meta-analysis of individual data in patients with acute severe bleeding (trauma and PPH): immediate treatment increased survival by more than 70% (OR 1.72; 1.42–2.10), each 15-minute delay reduced the survival benefit by 10%, and there was no benefit after three hours. For this reason, WHO recommends not starting tranexamic acid for the treatment of PPH more than three hours after birth (Ker 2024, Lancet; WHO).
Oxytocin infusion is titrated to a rate that controls atony: 40 U in 1 L of 0.9% NaCl is a common regimen; CMQCC recommends 10–30 U in 500 mL or 20–60 U in 1 L, given as a time-limited bolus over 10–15 minutes followed by maintenance at a slower rate, up to a maximum of 40 U/h. Doses up to 80 U in 500 mL over 30 minutes have been used but are not recommended—lower doses appear equally effective, and rapid infusion can cause severe hypotension and cardiovascular collapse. If prophylaxis was given as 10 U intramuscularly, that IM dose is not repeated; IM administration has a prolonged duration of action (at least one hour vs. 3–5 minutes IV). If bleeding persists, carboprost tromethamine and/or methylergonovine is added, with comparable efficacy (CMQCC 2022; UpToDate 2026).
| Medicine | Dose and route | Contraindications and caution |
|---|---|---|
| Carboprost tromethamine (15-methyl PGF 2α, Hebatate) | 0.25 mg IM every 15–90 minutes as needed; total cumulative 2 mg (eight doses). Approximately 75% of patients respond to a single dose — if there is no response after one or two doses, switch to another agent. It can also be injected into the myometrium (dilution of 0.25 mg in 20 mL 0.9% NaCl and aspiration before injection is recommended) | 🔴 Contraindicated in asthma. Nausea, vomiting, diarrhea and abdominal pain specific to prostaglandins are common. |
| Methylergonovine | 0.2 mg IM or directly into the myometrium—never IV. It can be repeated at intervals of 2–4 hours as necessary; If the first dose does not reduce bleeding, quickly switch to another agent | 🔴 Contraindicated in hypertension, coronary or cerebral artery disease and Raynaud's syndrome. It is not the agent of choice in the preeclamptic or chronic hypertensive patient. |
| Misoprostol (PGE 1 ) | The optimal dose is unclear and sources differ: the UpToDate author recommends 400 mcg sublingual, WHO recommends 800 mcg sublingual, CMQCC recommends 800 mcg sublingual or 600 mcg oral due to the risk of hyperthermia. The sublingual route reaches peak concentration in 30 minutes (~3 hours of effect); The rectal route is slower (up to one hour at peak) but longer-acting (four hours) and may be required in the unconscious patient. Vaginal route is not recommended | Useful in cases where injectable uterotonics are not available or contraindicated (hypertension, asthma). The meta-analysis did not find clear evidence that misoprostol is more effective than other uterotonics in the primary or adjunctive treatment of PPH. Fever ≥40 °C may occur: in one randomized trial, fever ≥38 °C was 58% versus 19%, fever ≥40 °C was 7% versus less than 1% in those receiving 600 mcg sublingual + standard uterotonics |
| Sulprostone (PGE 2 analogue; not licensed in the USA) | IV continuous infusion; initial 100 mcg/hour, increase to 500 mcg in the first hour, then 100 mcg/hour maintenance. Cumulative should not exceed 1500 mcg in 24 hours | Should be avoided or used with caution in patients with a history of asthma, glaucoma, severe liver or kidney failure. Serious cardiovascular events have been reported but are rare |
If oxytocin and tranexamic acid are ineffective, only partially effective, or there is a delay in administration, proceed quickly to intrauterine tamponade: intrauterine balloon, gauze tamponade, or low-level vacuum device that induces physiological uterine contraction. The choice is determined by local accessibility, clinician preference, and cost; ACOG published a clinical practice update on nonsurgical bleeding control devices in 2025. Uterotonics are continued until bleeding is controlled (ACOG Clinical Practice Update, 2025); Continued excessive bleeding indicates that the device is ineffective.
Uterine or hypogastric artery embolization is for hemodynamically and hemostatically stable patients who have not responded to less invasive treatments. In a review of twenty observational studies and 1739 patients, bleeding stopped completely with a single procedure in 89% of cases, re-embolization was required in 4%, and hysterectomy was performed in 7%; overall success is 90–97%, procedure-related morbidity is 3–6% (significantly lower than laparotomy). Gelfoam is the agent of choice; DIC is a risk factor for failure; Menstrual function and fertility usually return to baseline. The procedure may take 1–3 hours — laparotomy should be performed in the patient who is not stable enough to wait, and embolization should not be considered an emergency procedure for uncontrolled PPH of unknown cause (UpToDate 2026).
Laparotomy is indicated in the patient who has massive bleeding and is unstable after initial attempts. Vaginal postpartum atony is a last resort and rarely necessary; Since the abdomen is already open in cesarean section, surgical procedures are started earlier and these are successful in 85-90% of cases: uterine and utero-ovarian artery ligation, uterine compression sutures (B-Lynch and its derivatives) and hysterectomy when necessary. One-third of postpartum patients in shock require hysterectomy. Early resort to hysterectomy is appropriate in cases of extensive PAS or major uterine rupture; In atony, it is usually not the first attempt (UpToDate 2026).
Delaying measures provide time to resuscitate the patient. In external aortic compression, the performer positions himself on the epigastrium with his arms extended and applies firm pressure with a closed fist, slightly above and to the left of the umbilicus, in a way that presses the aorta against the vertebra. In REBOA, the balloon catheter placed through the deep femoral artery is advanced just below the renal arteries; Complete occlusion in the distal thoracic aorta should be limited to 15 minutes, and more distal occlusion should be limited to less than 30–60 minutes. Non-pneumatic antishock garment (NASG) can be applied in two minutes in environments where definitive treatment will be delayed; In an analysis of five observational studies, three of four studies reported a ≥50% reduction in median blood loss and a 38% reduction in total pooled mortality (UpToDate 2026).
Massive transfusion was defined as ≥10 U of whole blood or erythrocytes in 24 hours, ≥3 U of erythrocytes in 1 hour, or ≥4 blood components in 30 minutes; These are arbitrary cut-off points placed on a continuous blood loss. In massive bleeding, blood component replacement is more important than crystalloid—ongoing bleeding should be assumed in the patient whose blood pressure and urine output repeatedly fall despite reasonable volume replacement. There is no universally accepted guide to the empirical ratio: recommendations for the erythrocyte: fresh frozen plasma (FFP) ratio range from 1:1, 2:1, 3:2 and 6:4. If there are no laboratory results, bleeding continues, and the cause is atony or trauma, a reasonable approach is to give 4 U of erythrocytes followed by 4 U of FFP and maintain the 1:1 ratio until hemostasis tests guide targeted replacement; FFP is started earlier in cases of placental abruption or amniotic fluid embolism. One apheresis platelet suspension is given for every 4–6 U of erythrocytes (CMQCC 2022; UpToDate 2026).
| Parameter | Target | Note |
|---|---|---|
| Hemoglobin | >7.5 g/dL | A minimum of 8 g/dL after transfusion is recommended; Transfusion is rarely indicated when >10 g/dL |
| Platelet | >50,000/mm³ | <50,000/mm³, an apheresis platelet pack is given |
| Fibrinogen | >200 mg/dL | The normal level in pregnancy is 350–650 mg/dL (200–400 in non-pregnant women). In active bleeding, >300 mg/dL can be targeted; If <100 mg/dL, 10 U cryoprecipitate is given |
| Prothrombin time and aPTT | Less than 1.5 times control | If PT is longer than 1.5 times the control, 4 U FFP is given |
| Ionized calcium | >1 mmol/L (normal 1.1–1.3) | <1 mmol/L impairs coagulation and creates a risk of cardiac arrest. In massive transfusion, it should be measured every 15–30 minutes; For every 4 U of erythrocytes, 1 g of calcium chloride can be administered centrally over 2–5 minutes or 1–2 g of calcium gluconate can be administered IV over 2–3 minutes. |
| Potassium | Avoid hyperkalemia | In older units, supernatant K + increases from 2 mEq/L to ~45 mEq/L; Serum levels should be monitored serially when giving rapid and large numbers of units. |
| Core body temperature | ≥35.5 °C | All fluid and blood products should be warmed. The combination of hypothermia + acidosis increases the risk of clinically significant bleeding despite adequate replacement; If pH <7.1, bicarbonate is considered |
Fibrinogen is the “canary in the coal mine.” During PPH, fibrinogen falls to critical levels earlier than other clotting factors; It is therefore a more sensitive indicator of severe blood loss than prothrombin time, aPTT, or platelet count. If fibrinogen at diagnosis is <200 mg/dL, the positive predictive value for progression to severe PPH has been reported as 100%, and the negative predictive value for >400 mg/dL has been reported as 79%. Compared to those with fibrinogen >300 mg/dL, the likelihood of severe PPH increases approximately two-fold (OR 1.90; 1.16–3.09) in the 200–300 mg/dL range and 12-fold (OR 11.99; 2.56–56.06) in <200 mg/dL (UpToDate 2026).
Cryoprecipitate is the standard means of correcting fibrinogen deficiency; CMQCC recommends 30 units for fibrinogen <50 mg/dL, 20 units for <100 mg/dL, 10 units for 100–200 mg/dL, with each unit increasing plasma fibrinogen by at least 7–10 mg/dL. Fibrinogen concentrate (RiaSTAP, FIBRYGA) can be administered more quickly as it does not require thawing and is used when fibrinogen is <100 mg/dL and cryoprecipitate with FFP is not available; however, empirical use without coagulation test guidance should be avoided—in a randomized trial, administration of 3 g fibrinogen concentrate to PPH patients with baseline fibrinogen >300 mg/dL did not reduce blood loss, transfusion requirements, or postpartum anemia (UpToDate 2026).
A rare (approximately 1:1000) but critical presentation is acute obstetric coagulopathy: a disorder similar to that seen in amniotic fluid embolism with hyperfibrinolysis and acquired dysfibrinogenemia. In patients with very low fibrinogen and very high fibrin degradation products, large volume blood product transfusion may further dilute the remaining fibrinogen; If rapid measurement is not possible in this possibility (massive detachment, amniotic fluid embolism), early administration of high concentration lyophilized fibrinogen with tranexamic acid is recommended. Viscoelastic tests (TEG, ROTEM) provide holistic hemostasis assessment and targeted component selection at the bedside within minutes. Recombinant factor VIIa should be reserved for PPH and coagulopathy refractory to standard treatments (40–90 mcg/kg recommended for obstetric bleeding; start with low dose); For its effectiveness, at least platelets >50,000/mm³, fibrinogen >50–100 mg/dL, pH ≥7.2, and the absence of hypothermia and hypocalcemia are required (UpToDate 2026).
Retained placenta is diagnosed if the placenta is not expelled within 30 minutes after birth. Tissue is removed by hand or with ring forceps; Ultrasound is useful both in diagnosis and in guiding removal. If bleeding cannot be controlled, curettage is performed with a 16 mm aspiration catheter or preferably a wide blunt curette (banjo curette). Even if the placenta is manually removed, the myometrium in the placental bed may remain thinned and functionally abnormal; In a patient whose bleeding continues even though the placenta has been removed and reported to be complete, the lower segment persistently remaining loose or dilated while the upper segment contracts well should bring PAS to mind (UpToDate 2026).
In uterine inversion, a smooth round mass arising from the cervix or vagina is seen on vaginal examination, and the fundus is not in its normal position on abdominal examination. Inversion may be associated with fundal atony even if there is constriction in the lower uterine segment and cervix; The treatment is manual replacement of the uterus (UpToDate 2026).
Secondary (late) PPH occurs between 24 hours and 12 weeks after birth; The main causes are retained products of pregnancy, endometritis and subinvolution of the placental bed. The evidence base for management is weak: in a systematic review covering eleven studies and 834 participants, cessation of bleeding was achieved in 8.2–84.6% with medical therapy, 89.3–100% with surgical therapy, and 87.5–100% with transcatheter arterial embolization (71.9–73.7%, 89.3–92.0% and 87.5–100%, respectively, when limited to studies from the last 30 years). However, all of the studies were small, retrospective and observational and carry a risk of serious or critical bias; therefore, treatment selection depends largely on clinical context and local possibilities (Fox 2023, systematic review).
Maternal mortality varies widely worldwide: the mortality rate in patients who develop PPH ranges from 0.01% in the United Kingdom to 20% in some parts of Africa; It varies between 1 in 100,000 and 1 in 1000 per birth. Short-term morbidity headlines (UpToDate 2026):
In the long term, two conditions stand out. Sheehan syndrome (postpartum hypopituitarism) results from infarction of the pituitary gland — which enlarges during pregnancy — due to hypovolemic shock; the typical presentation is lactation failure combined with amenorrhea/oligomenorrhea, although any manifestation of hypopituitarism may appear from immediately postpartum to years later. A patient who remains hypotensive after PPH is controlled and volume replacement is completed should be evaluated in the early postpartum period for adrenal insufficiency; evaluation for other hormonal deficiencies may be deferred until 4–6 weeks later. In Asherman syndrome, approximately 90% of severe intrauterine adhesion cases result from uterine curettage performed for pregnancy complications such as PPH; uterine compression sutures have also been associated with adhesions. Before discharge, the patient should be counseled regarding normal vaginal bleeding, symptoms of excessive bleeding, and when to seek medical care (UpToDate 2026).
The risk of recurrence is significant: among patients with PPH in a prior pregnancy, recurrence reaches 18% in a subsequent pregnancy and 27% after two consecutive pregnancies complicated by PPH. More than 10% of patients who required transfusion for PPH due to uterine atony in their first pregnancy experienced recurrent transfusion-requiring atony in their next pregnancy. Risk depends in part on the underlying etiology: the risk of atony and retained placenta remains elevated across the next two vaginal deliveries, whereas the risk of laceration declines (UpToDate 2026).
Last updated: .