All 238 reports FDA received for the AMS 700 family during October 2026: the coded fields, plus a category assigned to each report from reading its narrative. That total includes 71 reports filed under a generic name with the model unknown, which are counted here as AMS 700s (section 2). The reports cover implants placed between 2006 and 2026 in 15 countries.
Please read this first. MAUDE counts reports, not failure rates. It does not record how many AMS 700s are implanted and working normally, so nothing below says how likely any problem is, and none of it can be used to compare the AMS 700 with another implant. Reports are unverified, most are written by the manufacturer from what a surgeon or company rep passed on, and one report is not always one man.
The short version
- 238 reports, no deaths. 230 (96.6%) are coded as hospitalization plus intervention, which here nearly always means revision or removal surgery.
- About 6 in 10 (142) were the device itself failing. Leaks and breaks lead: cylinders 41, tubing 28, fluid loss with no source identified 25, pump 18.
- Time in place and age. Taking the 142 device failures as one group, the device had been in for a median of 5 years, with a range of 0 to 12 (113 reports give both years; the middle half fell between 2 and 8 years). The men were a median of 68 years old, with a range of 36 to 87 (124 reports give an age; the middle half were 63 to 74). That describes the failures that were reported, not how long an implant lasts.
- About 2 in 10 (46) were position or fit: reservoir herniation 16, cylinders out of place 12, pump riding high 8, wrong size 8.
- About 1 in 8 (28) were infection or tissue breakdown: infection 17; erosion, extrusion or perforation 11.
- Early and late problems differ. Infection, pump position and sizing problems come up in the first year or two. Cylinder failures come late, at a median of 6.5 years.
- 32 reports involve the newer Tenacio pump, all implanted 2024-2026. Three are flagged by Boston Scientific as tied to its September 2025 Tenacio recall.
- Root cause is rarely confirmed. In 229 reports (96%) the device was not available for the manufacturer to examine. Only 3 include lab findings.
1. The batch at a glance
- Who filed: 236 by Boston Scientific, 2 by patients directly. A physician is named as the original reporter in 225, and 220 came through a company representative.
- Event type: 233 injury, 3 malfunction, 2 other. In MAUDE, "injury" includes anything that needed surgery to put right.
- When: 165 events (69%) are dated September or October 2026; 10 are dated before 2026. Median time from event to FDA receipt was 27 days.
- Where: 182 US (76.5%), 54 from 14 other countries, 2 not stated.
- Age: 32 to 87, median 68 (given in 212 reports). 69% were in their 60s or 70s; 16 men were under 50 and 16 were 80 or older.
2. Which products
- CX: 79 (54 with MS pump, 22 with Tenacio pump, 3 cylinders only)
- LGX: 42 (32 MS, 8 Tenacio, 2 cylinders only)
- CXR: 16 (13 cylinders only, 3 with MS pump)
- Model not identified: 72 (30.3%)
- Reservoir named as the device: 20 (16 Conceal, 4 spherical)
- Pump named as the device: 8 (7 MS, 1 Tenacio)
- "AMS 700 with Tenacio", cylinder type not given: 1 (a patient's report)
A caveat on those 72. 71 of them are filed under the generic name "AMS Inflatable Penile Prosthesis", usually because Boston Scientific was not told which product was involved. They are counted here as AMS 700s. That is the likely reading but not a confirmed one. 15 of the 71 narratives mention a reservoir or an AMS 700 pump, which points to a three-piece AMS 700. The other 56 mention neither, and a few of them could be the two-piece AMS Ambicor. Ambicor reports normally carry their own name, and FDA received 8 of those in October against 167 that name an AMS 700 or one of its parts. If the 56 follow the same mix, two or three of them are Ambicors. The FDA product code does not settle it: the 71 are coded FHW, as the Ambicor reports are, where reports that name an AMS 700 use JCW. The 72nd is a patient's own report that says AMS 700 without giving the model.
3. Why the surgery happened
FDA's codes are too coarse for this, so every narrative was read and given one main category: the chief reason for the operation or complaint. The category names are this analysis's own, not FDA's.
- Device malfunction or breakage: 142 (59.7%)
- Cylinder leak, hole, tear, aneurysm or layer failure: 41 (17.2%)
- Tubing crack, break, kink or detachment: 28 (11.8%)
- Fluid loss, source not found or not stated: 25 (10.5%)
- Pump malfunction: 18 (7.6%)
- Not working, cause not stated: 17 (7.1%)
- Auto-inflation, auto-deflation or won't stay inflated: 8 (3.4%)
- Reservoir leak, hole or pressure problem: 5 (2.1%)
- Position, migration or size: 46 (19.3%)
- Reservoir herniated, migrated or palpable: 16 (6.7%)
- Cylinder crossover, herniation, buckling or malposition: 12 (5.0%)
- Pump riding high or badly placed: 8 (3.4%)
- Wrong size: 8 (3.4%)
- Malposition, component not stated: 2 (0.8%)
- Infection or tissue breakdown: 28 (11.8%)
- Infection (including suspected, and wound breakdown): 17 (7.1%)
- Erosion, extrusion, perforation or fistula: 11 (4.6%)
- Other: 15 (6.3%)
- Pump hard to operate, or dexterity: 5 (2.1%)
- Pain or dissatisfaction with no device fault found: 5 (2.1%)
- Other clinical (antibiotic-coating allergy, Peyronie's, scar capsule round the pump): 3 (1.3%)
- Injury during the implant operation: 2 (0.8%)
- No narrative supplied: 7 (2.9%)
4. Device failures in detail (142)
Leaks and physical breaks (cylinders, tubing, reservoir, unlocated fluid loss) make up about seven in ten of these (99 of 142).
Time in place and age. Taking the 142 device failures as one group, the device had been in for a median of 5 years, with a range of 0 to 12 (113 reports give both years; the middle half fell between 2 and 8 years). The men were a median of 68 years old, with a range of 36 to 87 (124 reports give an age; the middle half were 63 to 74). By type of failure:
Code: Select all
Years in place
Type n Med Range
All 142 113 5 0-12
Cylinders 34 6.5 0-12
Tubing 25 4 0-10
Unlocated leak 22 5 0-10
Pump 13 3 0-10
Not working 8 2 0-12
Auto-infl/defl 6 0.5 0-4
Reservoir 5 8 0-9
Code: Select all
Patient age
Type n Med Range
All 142 124 68 36-87
Cylinders 37 69 38-87
Tubing 22 67.5 39-75
Unlocated leak 25 69 50-82
Pump 13 55 36-83
Not working 14 67.5 44-78
Auto-infl/defl 8 69.5 49-87
Reservoir 5 69 59-79
n = reports that give the figure. Years are explant year minus implant year, so 0 means removed in the same calendar year. Age is the age recorded on the report, not the age at implant. The 0 under cylinders is the likely data-entry error noted in section 7; the next shortest was 2 years.
Three cautions. These figures describe failures that were reported this month, not how long an AMS 700 lasts (section 7). Age does not mark out device failure: the median age of the whole batch is also 68. And the smaller groups (pump, auto-inflation, reservoir) rest on 5 to 13 reports, so their medians can move a lot.
Cylinders (41). 25 describe a hole, tear, rupture or leak. 14 describe the layers failing: the outer silicone layer splitting, peeling or shredding, fabric exposed or missing, or fluid between the layers. One is ballooning with no leak stated and one just says "defective". Six mention an aneurysm, bulge or ballooning. One failed set was examined in the manufacturer's lab (2124215-2026-57278, Spain): both cylinders had holes at the corners of folds and wear along the folds over their whole length. By model: CX 10, LGX 7, CXR 6, not identified 18.
Tubing (28). 11 cracks, 10 breaks, fractures or splits, 2 detached (one tube pulled out of the pump; one reservoir connector shown on MRI with no clear connection to the reservoir), 2 kinked or twisted, 1 leak at the cylinder tubing with no visible hole, 1 cut by another surgeon during a separate operation, 1 unspecified. Twelve were at or next to the pump, 9 on the cylinder tubing, 1 at the reservoir, 6 not located. Nine of the 11 cracks come from South Korea (section 10).
Fluid loss, source not identified (25). The system had lost fluid and was replaced. In 8 the surgeon looked and could not find the leak; the other 17 do not say where it was.
Pump (18). 7 not pumping or "not working", 3 with a flat, collapsed or dimpling bulb, 2 sticking, 2 with a hole or tear in the pump body, 1 that stopped inflating after the deflate button was pressed, 1 that would not deflate, and 2 that failed bench tests after return. Most complaints about the Tenacio pump fall under other headings and are covered in section 8.
Auto-inflation or won't stay inflated (8). 2 auto-inflating, 5 auto-deflating or losing rigidity, 1 doing both. In one case the removed pump worked properly when tested at surgery.
Reservoir faults (5). 2 holes, 1 fluid loss at the reservoir in an 8-year-old device, 1 found empty, 1 replaced for "excess pressure".
Not working, cause not stated (17). Includes two South Korean cases where "weak" cylinders were treated by adding saline with nothing removed, and two where air was found in the system but no hole.
5. Position and fit (46)
Reservoir (16). Herniated into the scrotum, groin or inguinal ring, or simply visible or palpable. Seven of the 15 with dates happened in the same or next calendar year after implant. Three narratives name a trigger: a cough, a sneeze, and lifting something heavy soon after surgery. Nine involved the flat Conceal reservoir, 3 the spherical one, 4 not stated. That is close to the mix in the batch as a whole (86 flat to 47 spherical where the type can be identified), so it does not single out either design. Ten of the 16 were fixed by replacing the reservoir alone.
Cylinders (12). 4 crossovers (3 confirmed, 1 suspected), 4 herniations through the corporotomy or out of the corpora, and 4 buckling, uneven or sitting short of the tip.
Pump (8). Riding high in the scrotum, migrated upward, or placed too close to the testicles. Seven of the 8 came up in the same or next calendar year. Five of the 8 are Tenacio systems.
Size (8). 3 too long (buckling, pain at the tip, "unstable"), 4 too short or not reaching the tip, 1 "incorrect size" with a floppy glans. One report from China attributes it to an inaccurate reading from the measuring tool. A ninth report, listed under erosion, says the wrong cylinder size may have contributed to cylinders and pump extruding in a 40-year-old; a smaller implant went in.
6. Infection and tissue breakdown (28)
Infection (17). 14 of the 16 with dates came out in the same calendar year as the implant. The two late ones were at 7 and 9 years, and the 9-year case was only suspected. Outcome: 7 received a malleable as a placeholder or replacement, 3 a new inflatable at the same operation after washout, 6 removal only (one with capped cylinders left in as spacers), 1 partial removal.
Erosion, extrusion, perforation, fistula (11). 3 extrusions through the skin (two pumps through the scrotum; one cylinders and pump), 4 cylinder erosions (one caught as "impending"), 3 perforations, 1 fistula. Two were in implants 18 and 20 years old. Both pump-through-scrotum cases were Tenacio systems implanted in 2026. Five more reports mention erosion or perforation as a secondary finding.
7. How long the devices had been in
197 reports give both an implant year and an explant year. MAUDE gives only the year, so "0" means removed in the same calendar year, and "1" means the following year, which could be anything from a few weeks to nearly two years.
Code: Select all
Years in place Reports Share
0 (same year) 56 28.4%
1 27 13.7%
2-3 29 14.7%
4-5 21 10.7%
6-7 25 12.7%
8-9 25 12.7%
10 or more 14 7.1%
Median: 3 years. This is not a lifespan estimate. It describes only the devices that generated a report this month. Implants that are still working never appear in MAUDE, so the table cannot say how long an AMS 700 lasts.
What it does show is that different problems arrive at different times:
Code: Select all
Problem n Med 0-1yr 5+yr
Cylinder failure 34 6.5 1 24
Unlocated leak 22 5 3 13
Tubing 25 4 7 11
Pump malfunction 13 3 4 3
Reservoir hernia 15 2 7 2
Erosion etc. 11 1 6 4
Wrong size 7 1 6 0
Pump position 8 0 7 0
Infection 16 0 14 2
n = reports with both years. Median is in years. "0-1yr" and "5+yr" are numbers of reports; "0-1yr" means removed in the same or the next calendar year.
The single early cylinder failure lists a 2014 manufacture date against a 2026 implant year, which looks like a data-entry error. Tubing splits into two groups: seven early cases (five of them Tenacio systems), and cracks and breaks in older devices, with cracks at a median of 6 years.
Of the 61 reports on devices implanted in 2026, only 15 are device failures. The rest are infection (13), reservoir herniation (6), cylinders out of place (6), pump position (5), erosion (4), sizing (3) and others. Of the 73 on devices implanted in 2021 or earlier, 59 (81%) are device failures, led by cylinders (24), unlocated fluid loss (13) and tubing (11).
8. The Tenacio pump and the recall
32 reports name a Tenacio system: 22 CX, 8 LGX, 1 pump on its own, 1 patient report. All but one (Canada) are from the US. All were implanted in 2024 (2), 2025 (14) or 2026 (16), and every one with dates came out in the same or the next calendar year. For a pump this new that is unavoidable, so the short time in place means little by itself. Four further reports, filed against the reservoir, list a Tenacio system as the companion device. They concern the reservoir (herniation, pressure, pain at the reservoir site) and are not counted here.
What the 32 describe:
- Pump riding high or badly positioned: 5
- Tubing broken, sheared, twisted or pulled out of the pump: 5
- Pump too hard to inflate or deflate: 4
- Auto-deflation or won't stay inflated: 3
- Won't inflate: 2 (one is the patient-filed report)
- Infection: 4
- Erosion or extrusion: 3
- Wrong size: 3
- Other: 3 (reservoir migration, crossover, Peyronie's)
In 10 of the 32 only the pump was changed, and in 4 of those the Tenacio was swapped for the older MS pump. Elsewhere in the batch the swap went the other way three times (MS replaced with Tenacio).
The recall. Boston Scientific began a Class II recall of specific Tenacio lots on 11 September 2025, and FDA posted it on 10 October 2025. FDA's notice gives the reason as "the potential for devices to experience inflation and/or deflation performance issues or difficulties" and the cause as a nonconforming material or component. For devices already implanted, the notice says nothing needs doing if the device works properly, and it recommends evaluation for anyone having trouble inflating or deflating.
Three reports in this batch carry the recall flag:
- 2124215-2026-52991 (Z-0142-2026): CX Tenacio pump migrated high, hard to inflate and deflate; swapped for an MS pump.
- 2124215-2026-55581 (Z-0142-2026): CX Tenacio pump hard to activate, confirmed by the surgeon; pump replaced.
- 2124215-2026-55840 (Z-0144-2026): LGX Tenacio, pain and would not inflate; no surgery described yet.
A check of this batch's lot numbers against FDA's published lot lists turned up two more reports whose lots are on the Z-0142-2026 list but which are not flagged: 2124215-2026-53401 (patient struggling to operate the device) and 2124215-2026-54564 (removed for being too short). None of the other Tenacio lot numbers in the batch appeared on the lists checked.
Manufacture dates add something. All four "too hard to operate" reports and two of the three auto-deflation reports involve pumps built before the recall date. Of the 13 Tenacio devices built on or after it, one has an inflation or deflation complaint; the rest are infection (4), tubing (3), pump position (2), erosion (2) and Peyronie's (1).
What this cannot tell you: whether the Tenacio has more problems than the MS pump. For devices implanted in 2025-2026 the batch holds 30 Tenacio reports and 38 MS-pump reports, but without knowing how many of each were implanted those numbers cannot be turned into a comparison. A recalled product also attracts more reporting.
Separately, and outside this batch: FDA posted a small Class II recall on 9 November 2026 (Z-0396-2027, 7 units) because some packages labelled as 0.5 cm rear tip extenders may contain 1.5 cm ones.
9. What the surgeons did
- Whole device (or simply "the device") replaced with a new inflatable: 104 (43.7%)
- Cylinders and pump replaced, reservoir left in: 21 (8.8%)
- Pump only replaced: 21 (8.8%)
- Reservoir only replaced: 14 (5.9%)
- Other partial swaps (cylinders or rear tip extenders only; reservoir plus pump): 6 (2.5%)
- Converted to a malleable: 21 (8.8%), 16 of them naming the Tactra
- Removed, no replacement described: 22 (9.2%)
- Replaced with another maker's implant: 2 (0.8%)
- Fixed without removing anything (repositioned, corporotomy repaired, saline added): 7 (2.9%)
- Still implanted, or no surgery described: 8 (3.4%)
- Not stated: 12 (5.0%)
10. Where the reports came from
US (182), from 35 states and DC: Texas 28, Florida 19, California 19, North Carolina 13, Ohio 8, South Carolina 7, Illinois 7. That most likely tracks population and implant volume.
Other countries (54): South Korea 23, China 9, Brazil 5, Canada 3, Italy 2, India 2, Spain 2, Honduras 2, and one each from the UK, Australia, Panama, Israel, Estonia and Russia.
South Korea stands out. All 23 reports are mechanical: 10 tubing (9 of them a crack in the pump or cylinder "connecting tube"), 9 cylinder holes, ballooning or defects, 2 "weak" cylinders topped up with saline, 1 dimpling pump, 1 unspecified. The cracked-tube devices had been in for 2 to 9 years, with a median of 8. Twenty of the 23 list separately packaged components (CXR, CX or LGX cylinders, or the MS pump) and 13 came through a distributor.
Reports from outside the US skew mechanical: 40 of 54 (74%) are device failures against 100 of 182 (55%) in the US, and only 3 of the 54 are infection or erosion. That is more likely a difference in what gets reported than a difference in outcomes.
11. FDA's own codes
Device problem codes (a report can carry several): Mechanical Problem 71, Fluid/Blood Leak 68, Material Puncture/Hole 61, Inflation Problem 53, Migration 35, Adverse Event Without Identified Device or Use Problem 27, Inadequacy of Device Shape and/or Size 11, Collapse 11, Air/Gas in Device 11, Degraded 10, Deflation Problem 9, Unintended Deflation 8, Nonstandard Device 8, Malposition of Device 7.
Health effect codes: No Clinical Signs, Symptoms or Conditions 158 (66.4%), Pain 20, Unspecified Infection 15, Unspecified Tissue Injury 13, Discomfort 11, Erosion 11, Deformity/Disfigurement 8, Swelling/Edema 6, Perforation 5, Capsular Contracture 3.
"No clinical signs" does not mean nothing happened. 155 of those 158 reports are still coded as hospitalization plus intervention.
12. How solid is the data
- Product often unknown. 72 reports (30.3%) have no catalog number and 74 have no lot number.
- Devices rarely examined. 229 (96.2%) were not available to the manufacturer, 7 list a return date, and 3 have lab results: worn-through cylinders (above), an MS pump that failed deflation and activation tests ("component failure"), and a pump that passed its leak test but failed its functional test ("quality control deficiency"). Four more were returned in October and have no results yet. Nearly everything known about cause is the surgeon's description at revision.
- Tick-boxes understate device problems. The "product problem" box is ticked on 21 reports (8.8%). Of the 215 where it is not, 92 carry a leak or hole code.
- Dates are soft. 43 event dates fall on the 1st of a month and 8 reports say the date was estimated.
- Still moving. 33 reports (13.9%) already include follow-up information, and more will be amended.
- No confirmed duplicates, but the check is weak outside the US. Among the 172 reports that give both a state and an age, no two share a state, age and event date, and no lot number appears twice. That test catches little abroad, where 53 of the 54 reports give no state and 18 give no age. Two South Korean reports (2124215-2026-56474 and 2124215-2026-56210) share an event date, give no age, and both describe a loosely worded fault involving the pump, so they may be one case. Two Honduras reports two weeks apart, both for a 39-year-old (a replacement, then an infection), may be one man.
13. Individual reports worth reading
- MW6003279 (patient-filed, age 74): Tenacio pump stopped inflating. The correspondence quoted in the report shows the company telling him the warranty runs through his surgeon, and that it could not commit to a free replacement until it had the surgeon's paperwork and had inspected the device.
- MW6003357 (patient-filed, age 62): pump bulb collapsed and would not refill about two weeks after he started using it. A second surgeon replaced it with a Coloplast Titan. He kept the removed device and says the company declined to examine it because it had not come through the proper chain of custody.
- 2124215-2026-56481 (China, age 48): bleeding after implant; emergency surgery found a ruptured bladder and the reservoir inside it. He had a previous bladder repair the surgeon had not been told about.
- 2124215-2026-56274 (Georgia, age 51): bladder injured while placing the reservoir; the surgeon switched to a Tactra malleable.
- 2124215-2026-52662 (Wisconsin, age 60): urethra perforated during dilation at the original implant, followed by infection and removal.
- 2124215-2026-57391 (Texas, age 65): allergic reaction to the InhibiZone antibiotic coating in a man with an undisclosed tetracycline allergy; replaced with uncoated components.
- 2124215-2026-57685 (New York): tubing cut by another surgeon during a separate operation; the reservoir had to be replaced.
- 2124215-2026-57367: an MS pump that apparently failed during an implant procedure (a second device of the same model was used to finish). On the bench it failed deflation and activation tests.
- 2124215-2026-57644 (California, from a legal claim): device stayed semi-inflated for 24 hours several weeks after implant; multiple revisions.
14. Practical points
- Tenacio owners: the recall is lot-specific. If the pump has become hard to inflate or deflate, FDA's notice recommends getting it evaluated, and the surgeon's office or Boston Scientific can check the lot. If it works properly the notice says no action is needed.
- Reservoir herniation often shows up early. Seven of the 15 dated reports came in the same or the next calendar year after implant, though the median was 2 years. Three reports tie it to a cough, a sneeze or heavy lifting soon after surgery, which fits the usual advice to avoid heavy lifting early on.
- Tell the implant surgeon about antibiotic allergies and earlier pelvic surgery. InhibiZone contains rifampin and minocycline (a tetracycline). A previous bladder repair and an undisclosed allergy each led to a second operation in this batch.
- Tell any other surgeon working in the groin, pelvis or scrotum that there is an implant. One man's tubing was cut by another surgeon during a separate operation.
- Old devices mostly fail by leaking. Of the 72 devices that had been in for 5 years or more, 48 came out for a cylinder failure, unlocated fluid loss or a tubing fault (mostly cracks and breaks). Surgeons often kept the old reservoir.
- Want the manufacturer to examine a failed device? Raise it with the surgeon before the revision. One patient who kept his own explant reports being told it could not be analysed.
15. Method and limits
- Source: the October 2026 AMS 700 export of MAUDE reports, 238 reports with 104 fields each. Coded fields were counted as filed.
- Scope: the export includes 71 reports filed under a generic product name with the model unknown. They are counted as AMS 700s, which is likely but not confirmed (section 2).
- Each narrative was assigned one main category. Where a report describes two problems (for example a herniated reservoir plus a ruptured cylinder layer) it sits under the one named first or most clearly behind the surgery. Another reader could reasonably move a handful, which would shift subtotals by a point or two.
- The 7 reports with no narrative are kept separate. Their FDA codes: fluid leak 2, inflation or mechanical problem 4, device contamination 1.
- Country codes were read as FIPS-style codes, the only reading under which all of them are valid (KS = South Korea, CH = China, SP = Spain, AS = Australia).
- Recall details are from FDA's recall database: Z-0142-2026, Z-0144-2026, Z-0396-2027. Check a lot number against those records rather than relying on the matches reported here.
- FDA's own description of what MAUDE can and cannot show is on its website, on the page titled "About Manufacturer and User Facility Device Experience (MAUDE) Database".
