Cardiosave balloon pump: 19,898 MAUDE reports and an FDA warning to switch devices

#19,898 Reports and Climbing: The Cardiosave Adverse Event Signal

The FDA's MAUDE database has accumulated 19,898 adverse event reports across the Cardiosave intra-aortic balloon pump family since 2019. Of those, 82 carry an event type of death and 77 an event type of injury. These reports arrive in a passive surveillance system: de-identified, unverified, and counted as reports rather than rates. A death code does not establish that the device caused the death, and MAUDE holds no denominator for how many devices were in service in any given period.

The reports have not tapered. Annual volume climbed from 1,205 in 2019 to a peak of 3,840 in 2025, and our pipeline flagged an increasing trend in Type B Plug reports as an active critical signal.

Year	Reports	Deaths
2019	1,205	7
2020	1,329	14
2021	1,779	15
2022	2,211	6
2023	3,612	13
2024	3,473	5
2025	3,840	12
2026	2,449	10

The FDA has issued formal warnings on this device family, including a Letter to Health Care Providers about continued safety and quality concerns and an earlier update on device failures. We have written before about why signal detection in post-market surveillance is broken, and the Cardiosave record is a case study in what that failure looks like over time. You can query the underlying data yourself through the public MAUDE lookup tool.

#What Is the Cardiosave Hybrid? Device Approval and Market History

Datascope Corp., a subsidiary of MAQUET Cardiovascular, manufactures the Cardiosave Intra-Aortic Balloon Pump. MAQUET received FDA 510(k) clearance and CE mark for the device in November 2011, announced at the TCT meeting in San Francisco, with US commercial availability starting January 2012. Datascope filed a later 510(k), K181122, on April 30, 2018, and the FDA returned a Substantially Equivalent decision on May 31, 2018.

Clinicians use the Cardiosave for hemodynamic support in cardiogenic shock and high-risk percutaneous coronary intervention. The device family spans multiple plug variants, and the report burden splits unevenly across them.

Component	Reports
CARDIOSAVE HYBRID, TYPE B PLUG	13,331
CARDIOSAVE HYBRID W/ E/F PLUG	1,993
CARDIOSAVE HYBRID, TYPE I PLUG	1,409
CARDIOSAVE HYBRID, TYPE G PLUG	1,386
CARDIOSAVE HYBRID, 3.1 EDITION	779
CARDIOSAVE HYBRID TYPE D PLUG	272
CARDIOSAVE HYBRID TYPE I (AU)	266
CARDIOSAVE RESCUE	253
CARDIOSAVE RESCUE, 3.1 EDITION	57
CARDIOSAVE HYBRID TYPE J PLUG	52

The full record sits on the device detail page.

#Component-Level Breakdown: Type B Plug Dominates Adverse Reports

The Type B Plug carries 13,331 of the 19,898 reports in this device family, 67 percent of the total. Every other variant trails by an order of magnitude: the W/E/F Plug at 1,993, the Type I Plug at 1,409, and the Type G Plug at 1,386. Rescue variants log 253 reports, and the regional Type I (AU) version 266.

That concentration is worth sitting with. When one component out of ten accounts for two-thirds of the reports, the pattern points toward a specific component or design factor rather than a uniform, family-wide fault. It does not by itself identify the mechanism, and reporting patterns can reflect how many of each variant are in the field, which MAUDE does not tell us.

#Top Failure Modes: Gas Leaks, Display Problems, and Unexpected Shutdowns

Gas or air leak is the single most reported product problem at 3,023 reports. Inability to auto-fill follows at 1,150, and break at 1,111. Display or visual feedback problems account for 890 reports and overheating of the device for 875.

Unexpected shutdown appears 816 times, and it maps onto the theme the FDA identified across the recalls: a common thread was that the devices would unexpectedly stop working. Our narrative review grouped reports into failure-mode clusters. The largest were gas loss alarms with patient coughing (101), overheating alarms during use (97), and malfunctions with an over-temperature issue (96).

Failure mode	Reports	Emerging
Gas Loss Alarm with Patient Coughing	101	no
Overheating Alarm During Use	97	no
Malfunction with Over Temperature Issue	96	no
Defective Display During Routine Check	79	no
Balloon Rupture After Catheter Insertion	70	no
Screen Flickering During Routine Check	67	no
Autofill Failure During Use	62	no
Battery Failure During Unplugging	54	no
Balloon Damage During Catheter Insertion	54	no
Battery Failure During Operation	50	no

The balloon-related clusters carry direct clinical stakes. Balloon rupture after catheter insertion (70 reports) and balloon damage during insertion (54) sit alongside the same mechanism the FDA cited in a January 2023 recall, where a compromised balloon could allow blood into the pump and produce helium gas emboli that damage organs including the brain.

#Patient Outcomes in MAUDE: What the Data Show and Don't Show

Here the data works against its own headline. Of the 19,898 reports, 16,498 (83 percent) carry the patient-problem code "No Clinical Signs, Symptoms or Conditions." Another 2,523 cite insufficient information.

Patient problem	Reports
No Clinical Signs, Symptoms or Conditions	16,498
Insufficient Information	2,523
No Patient Involvement	1,533
No Consequences Or Impact To Patient	798
No Known Impact Or Consequence To Patient	575
Appropriate Term / Code Not Available	122
Death	24
Low Blood Pressure/ Hypotension	24
Cardiac Arrest	24

Cardiac arrest, hypotension, and death each appear in only 24 reports at the patient-problem level. That number reads low against 82 death event-type codes, and the gap is a feature of how MAUDE is built rather than evidence of safety. The system is passive and voluntary, and it captures device malfunctions far better than it captures downstream clinical harm. We break down the mechanics in how to search the FDA MAUDE database, and the broader structural problem in post-market surveillance in the United States.

#FDA Actions and Recalls: A Cascade of Safety Alerts Since 2023

Recalls are a separate FDA system from MAUDE, and the recall record here is dense. From January 2023 through April 2024, Getinge and Datascope initiated 12 voluntary recalls, 8 of them Class I, the most serious category.

Widening the window changes the scale. Querying FDA recall records for Cardiosave products returns 57 recall entries dating back to March 2012, 30 of them Class I. Fifty-two carry a status of open and classified, including the most recent, filed in November 2025. A Class I designation means the FDA judged that use of the product could cause serious injury or death. These counts run at the recall-record level, and a single corrective action often generates several entries covering different models and part numbers, so 57 entries do not mean 57 distinct events. What the pattern does show is a device family that has stayed in service through more than a decade of corrections, most of which the FDA has not yet closed out.

Recall	Class	Date	Reason
Z-0916-2026	Class II	2025-11-04	IFU addendum revises Preventative Maintenance schedule per June 2023 Service Manual
Z-0917-2026	Class II	2025-11-04	IFU addendum revises Preventative Maintenance schedule per June 2023 Service Manual
Z-0932-2026	Class III	2025-10-23	IFU addendum updates Vibration and Shock Table standards
Z-0931-2026	Class III	2025-10-23	IFU addendum updates Vibration and Shock Table standards
Z-3136-2024	Class I	2024-08-08	Software correction (Rev D.01) for risk of hemodynamic instability and failure
Z-3137-2024	Class I	2024-08-08	Software correction (Rev D.01) for risk of hemodynamic instability and failure
Z-3135-2024	Class I	2024-08-08	Software correction (Rev D.00) for risk of hemodynamic instability and failure
Z-3134-2024	Class I	2024-08-08	Software correction (Rev D.00) for risk of hemodynamic instability and failure
Z-2431-2023	Class I	2023-07-31	Device not charging; console not fully inserted in cart
Z-2486-2023	Class II	2023-07-31	IAB Fiber Optic Sensor input failure on connector insertion

The four August 2024 Class I recalls (Z-3134 through Z-3137) addressed software problems posing a risk of hemodynamic instability and device failure. The January 2023 balloon recall cited the helium gas emboli mechanism. A separate Class I recall for System Over Temperature alarms causing pump stops rested on 213 complaints between January 2021 and June 2023, including four injuries and one death. The July 2023 pair split between a charging failure tied to improper cart docking (Z-2431) and a fiber optic sensor failure (Z-2486).

Two events outside the MAUDE record define the regulatory posture. On May 8, 2024, the FDA issued a Letter to Health Care Providers recommending hospitals transition away from Cardiosave devices and seek alternatives. Two months earlier, on March 1, 2024, TÜV SÜD, the EU notified body, temporarily suspended the device's CE certificate.

#Competitive Context: Cardiosave vs. Impella in Mechanical Circulatory Support

The Cardiosave's main clinical competitor is the Impella device from Abiomed, now part of J&J MedTech, which received FDA 510(k) clearance in June 2008. Comparative analyses have shown broader revascularization potential with Impella in high-risk PCI relative to traditional balloon-pump support.

The larger question sits above the device choice. The IABP-SHOCK II randomized trial enrolled 600 patients between 2009 and 2012 and found that IABP did not reduce 12-month all-cause mortality in acute myocardial infarction complicated by cardiogenic shock. Twelve-month mortality reached 52 percent in the IABP arm versus 51 percent in control, a relative risk of 1.01 with a p-value of 0.91. The New England Journal of Medicine published the trial in 2012 and The Lancet in 2013, and the efficacy questions it raised reach past any single manufacturer.

#What MAUDE Can and Cannot Tell Us: Surveillance Limits and Interpretation

MAUDE is passive and voluntary, and its numbers lag reality. The FDA disclosed that it has monitored Cardiosave shutdowns in MDRs since 2017, more than seven years before the recall cascade began in earnest in 2023. A signal sat in the record for the better part of a decade.

The scale of the reporting stream shows in the FDA's own count. In the 12 months preceding the May 2024 safety letter, the agency received 2,964 MDRs related to Cardiosave IABPs, of which 15 were reported as serious injury or death. That ratio, thousands of malfunction reports to a small handful of coded harms, echoes what we found in our own query, where 83 percent of reports carry no recorded clinical sign.

Two readings of that pattern are possible, and both are partly true. The device fails often, and most failures are caught before a patient is harmed. And MAUDE undercounts harm, because clinicians often attribute a bad outcome in cardiogenic shock to the underlying disease rather than the pump. For rare, severe outcomes like death and cardiac arrest, the second effect dominates. MAUDE counts reports, not unique patients, and it holds no installed-base denominator, so no failure rate can be computed from it.

This is the same structural gap we traced in our analysis of why signal detection is broken and in the HVAD adverse-event record, where reports kept arriving long after the device left the market. The lesson repeats: a surveillance system that measures reports well and harm poorly will flag the noise before it confirms the injury.

#For Hospitals and Clinicians: Practical Guidance and Next Steps

The FDA's May 2024 letter is explicit: transition away from these devices and seek alternatives, if possible. The failure modes in this analysis, unexpected shutdown, gas leak, overheating, and auto-fill failure, occur during active hemodynamic support, where an interruption risks collapse.

Facilities still running Cardiosave units should align preventative maintenance with the June 2023 Service Manual update referenced in the November 2025 Class II recalls Z-0916 and Z-0917. Watch for the signatures this data surfaces: gas loss alarms, display problems and screen flickering during routine checks, overheating alarms, and battery failures during operation or unplugging. Document every malfunction, even when no injury is evident, because the reports that look inconsequential today build the trend line that eventually forces action.

#Methodology: How These Numbers Were Derived and What They Mean

All adverse-event figures come from FDA MAUDE data, accessed through the Claripulse Data Pack for the Cardiosave intra-aortic balloon pump family, and spanning January 2019 through the most recent available records in 2026. Totals include all device problem codes, patient problem codes, and event types submitted by hospitals, manufacturers, and users. We derived the component breakdown by filtering MAUDE records on product name and code.

We built the failure-mode clusters from narrative fields using automated grouping followed by manual review, and we flag emerging clusters when the trend is increasing. We matched recall and safety-letter references to FDA recall records and safety communications. External comparative data draws on the IABP-SHOCK II trial, FDA regulatory records including K181122, and manufacturer disclosures.

This analysis maps a landscape. It does not establish causation. MAUDE reports are not rates, an event-type death code is not a finding that the device caused a death, and the record is de-identified. For the mechanics behind these limits, see how to search the FDA MAUDE database, and to work with the underlying records, use the device detail page.

This analysis is based on publicly available FDA MAUDE data and published literature. It does not constitute medical advice. MAUDE reports are submitted under FDA regulation, are de-identified, and may be incomplete or unverified. No attempt should be made to re-identify any patient.