The six wrong-site procedures in Pennsylvania operating suites this quarter were an all-time low since reporting of patient safety events began in mid-2004 (see Figure). Two of the procedures were wrong-side anesthesia blocks. Two were wrong-side pain blocks. One was a wrong-side needle localization of a breast lesion for a breast biopsy. One was the removal of a wrong skin lesion. Going into 2013, 88 days passed since a surgeon had operated on the wrong site in the operating room (OR) and 20 days passed since an anesthesiologist had blocked the wrong site. Figure. Pennsylvania Patient Safety Authority Wrong-Site Surgery Reports by Academic Year
Some near-miss reports from this quarter illustrate the importance of using the time-out as a final check for information errors:
The CRNA [certified registered nurse anesthetist] gave the report to the circulator stating consents were signed and the patient was marked. The patient stated right ear surgery when asked upon entering the OR, and the right ear was marked. The history and physical noted right ear surgery. When reading the OR consent [during the] time-out, the RN [registered nurse] noticed no right or left ear was noted in the surgery description. The time-out stopped until [the ear was] clarified. The surgeon agreed it was the right ear.
The patient was on the OR schedule for a left inguinal hernia. The circulating nurse opened the electronic patient record in preparation for the procedure and used the information for this patient to complete the information on the time-out board. The circulating nurse retrieved the patient from the preoperative area, confirming the patient’s identity with the parents, the chart, and the patient bracelet. The circulating nurse entered the OR with the patient and proceeded to perform a time-out, identifying the patient, birth date, weight, allergies, and procedure. During the time-out procedure, the nurse anesthetist asked why the patient’s first name on the time-out board was different from the patient bracelet and patient chart. The surgeon stated the patient was a twin and the twin had been scheduled instead incorrectly.
Another report illustrates the struggle facilities have ensuring compliance with best practices to prevent wrong-site surgery with the time-out:
The doctor inserted the cystoscope prior to the time-out. When reminded, he continued the procedure while doing the time-out. [He was] reminded again of the need to wait to start the procedure until after the time-out, and the doctor continued talking.
Recently, Mehtsun et al.1 reviewed surgical “never events” in the National Practitioners Data Bank from September 1, 1990, through September 30, 2010. For wrong-site, wrong-procedure, and wrong-patient events, they identified an average payment of $179,575, in 2010 US dollars, for each of an average of 250 events with payments per year. They extrapolated a total of 2,058 wrong-site, wrong-procedure, or wrong-patient occurrences per year in the United States. When outcomes were described for events resulting in payments, 8% resulted in deaths, 48% in permanent injury, 42% in temporary injury, and 2% in emotional injury.
An Analysis of 500 Wrong-Site Surgeries
The first 500 wrong-site surgery events reported to the Pennsylvania Patient Safety Authority between July 2004 and August 2012 were reviewed for the relationship between the type of wrong-site event, the procedure intended, and the location of the intended incision. The primary intent was to answer the question: What sites benefit from marking and having the mark referenced during the time-out?
Of the 500 events, 9 (1.8%) were OR procedures done on the wrong patient, approximately 1 per year. However, 32 months passed since the last report of an OR procedure on the wrong patient in Pennsylvania.
Of the 9, 8 involved the use of incorrect information, as follows:
- 4 involved patients with the same or similar names
- 2 involved another patient’s information entered in the chart
- 1 involved using another patient’s chart
- 1 involved operating on patients out of the scheduled order
Two reports documented suboptimal patient identification practices:
The physician entered and identified the patient by first and last name. The patient responded affirmatively. . . The RN entered and identified the patient by name, and the patient responded affirmatively. . . . The RN spelled the patient’s last name, and the patient responded affirmatively.
The patients were coming back to the OR out of order. . . . His name bracelet was not checked, and he was not asked to identify himself.
However, the rarity of wrong-patient events in the OR is consistent with good compliance with verification using the patient’s statements of two identifiers. This “active voice” verification of the patient’s identity has been observed during site visits at multiple facilities involved in collaborations to prevent wrong-site surgery.
Physicians initiated a wrong procedure during 58 of the 500 events (11.6%), with 42 of the 58 wrong procedures (72.4%) being completed. The types of procedures were as follows:
- 14 hand procedures
- 12 insertions of an incorrect device, including 1 incorrect type of ear tubes
- 10 ear, nose, and throat (ENT) procedures, including the previously noted event involving ear tubes
- 5 eye procedures
- 4 gynecological procedures
- 3 breast procedures
- 3 orthopedic procedures
- 3 urological procedures
- 2 pain management procedures
- Plus a bariatric procedure, a hernia repair, and a wrong endoscopy
The wrong-hand procedures exhibited an interesting pattern: 12 of the 14 were incisions for unintended carpal tunnel release. For 9 of the 12, the intended procedure was a trigger finger release, not a carpal tunnel release, exactly the same situation described in a case report in the New England Journal of Medicine.2
Of the 58 wrong procedures, 9 were additional procedures to those that were intended. Of those, 3 were unintended tonsillectomies in addition to planned adenoidectomies, 2 were unintended bilateral salpingo-oophorectomies in addition to planned hysterectomies, and 1 was a bilateral myringotomy instead of a unilateral myringotomy.
The tendency to do carpal tunnel releases instead of trigger finger releases, tonsillectomies with adenoidectomies, and salpingo-oophorectomies with hysterectomies suggests problems of automatic thinking by the surgeons and the need for OR staff to maintain situational awareness of the intended procedures throughout all operations:
The patient was in the OR for bilateral myringotomy with insertion of PE [pressure equalization] tubes and adenoidectomy. The patient was draped, and a time-out was performed prior to beginning the myringotomy. When the bilateral myringotomy was completed, the surgeon placed the mouth gag. The scrub nurse set up the coblator, and, when she looked inside the mouth, she saw the surgeon was removing the right tonsil. [The scrub nurse and surgeon] both noted at that time the patient was for adenoidectomy.
The patient was scheduled for a right trigger thumb release. The surgeon made the incision for carpal tunnel release. The surgical tech questioned the procedure.
The patient was scheduled for left trigger thumb release. . . . The patient held up her marked thumb, which was initialed by [the doctor]. . . . The patient was prepped and draped. The time-out was announced-—the patient name and proper procedure [were verified]. [The doctor] began injecting the palm as though he was doing a carpal tunnel instead of a trigger thumb release. The CRNA asked [the doctor] if he was injecting for a carpal tunnel instead of a trigger thumb. [The doctor] stopped.
Some of the unintended carpal tunnel releases were facilitated by the absence of a site mark or time-out:
Scheduled for release of trigger finger; consent indicated same; site marked by surgeon; . . . during prep, site mark washed off with alcohol; MD proceeded to do carpal tunnel, then realized he was to do trigger finger; MD told staff he was thinking about a patient he had done previous day; MD said the time-out was done.
Time-out was not performed. Incision made for left carpal tunnel. Surgeon realized patient was to have release of trigger finger.
Patient was scheduled for a trigger finger release. . . . No time-out process was completed. The surgeon started the procedure and made an incision as if the patient were undergoing a carpel tunnel release. The surgeon immediately realized his error.
Physicians initiated an intended procedure at an incorrect site on the correct patient according to 433 of the 500 event reports (86.6%). The errors in the location of the incisions or punctures were subdivided as follows:
In three reports, the information was insufficient to make such a determination.
Counting the report of a procedure started both on the wrong side and at the wrong level, 307 OR procedures were initiated on the wrong side of body. Wrong-side procedures constituted the vast majority (61.3%) of all 500 wrong-site surgery reports. They represented 71.3% of all intended procedures done on the correct patient but initiated at the wrong location on the body among the 430 patients for whom the location could be determined.
The exact anatomic location of the surgery was described for 283 of the 307 wrong-side procedures and is presented in Table 1. For some of the events, the wrong-side error did not occur during the primary procedure but during the anesthesia block or the tissue donor site procedure. Wrong-side anesthesia blocks done by anesthesiologists accounted for 104 of the 307 wrong-side procedures in the OR (33.8%), and wrong-side pain procedures accounted for another 41 (13.4%). Surgical procedures accounted for slightly more than half (52.6%).
|No. Of Wrong-Side |
|Chest or lungs||7|
|* 5 of 7 craniotomies were emergencies for brain injuries|
† 1 of 6 hip procedures was for a hip fracture
Of the 43 wrong-side spinal procedures, 29 (67.4%) were injections to control pain.
The need for intraoperative verification of ureteral stent placement with imaging studies has been discussed in a previous issue of the Pennsylvania Patient Safety Advisory.3
The only wrong-site emergency trauma procedures done in the OR were wrong-side craniotomies, as illustrated by the following:
There was no consent obtained for this emergent case. The patient arrived intubated in the OR. The OR staff was set up for a right craniotomy, but the surgeon was positioning the patient for a left craniotomy. The OR staff assumed amongst the rush of the case that the case was booked wrong and that it was a left-side craniotomy. [After making the incision, the surgeon] recognized the operation was on the wrong side. The procedure was immediately stopped. The dura was not opened. The incision was closed and dressed, and the patient was repositioned for a right craniotomy. . . . No time-out [had been] done.
A fracture by itself is not always adequate to indicate the correct site for surgery, as illustrated by the following:
Surgery was scheduled and consent obtained for repair of a right hip fracture. The patient marked the site, and team verification “pause” occurred. However, the patient was positioned with the left hip draped and prepped, and the surgery proceeded. After the incision, the error was realized. The incision was sutured and the patient repositioned, and surgery resumed on the right hip.
Counting the report of the procedure started both on the wrong side and at the wrong level, 65 OR procedures were initiated on the level of a multitiered skeletal structure. Wrong-level procedures constituted 12.9% of all 500 wrong-site surgery reports. They represented 15.0% of all intended procedures done on the correct patient but initiated at the wrong location on the body among the 430 patients for whom the location could be determined.
All but one of the wrong-level events was a procedure done at the wrong spinal level. One event of the 500 was the resection of the second rib, instead of the first rib, on the correct side. For events requiring intraoperative verification of the correct level of a skeletal structure by imaging studies, the event is not considered a wrong-level event if a correct adjustment is made in the location in response to the intraoperative verification before the definitive part of the procedure is initiated.
Of the 64 wrong-level spinal procedures, one event described a patient who had a lumbar procedure instead of a cervical procedure:
The patient was scheduled for an epidural steroid injection. . . . The time-out was done using the consent. However, the site had not been marked in the preoperative area and the consent did not specify a site. The patient stated only that he was having an epidural steroid injection. The doctor stated during the time-out that the lumbar area was the correct area. After the lumbar procedure, the CRNA discovered “cervical” on the medical record and informed the physician.
Of the 64 reports of wrong-level spinal procedures, 57 (89.1%) were open procedures and 7 (10.9%) were pain procedures. Open spinal procedures were more likely to be done at the wrong level (56) than on the wrong side (13), not considering the procedure done at both the wrong level and on the wrong side.
Of the 64, 6 might have been due to preoperative information errors and 56 were most likely due to intraoperative misperceptions; the sources of error were ambiguous for 2. The intended and erroneous vertebral bodies used to verify the location of the procedure were mentioned in 53 reports. Most of the vertebrae intended to be used for intraoperative verification of the spinal procedure were in the lumbosacral region of the spine (55.7%), followed by the cervical region (30.2%) and the thoracic region (14.2%). For cervical and upper thoracic spine procedures, there was a tendency to erroneously identify vertebrae that were more caudad (closer to the coccyx). For lower thoracic and lumbosacral spine procedures, there was a tendency to erroneously identify vertebrae that were more cephalad (closer to the head) (see Table 2).
|Errors Caudad‡||Errors |
|C1||0|| || || |
|C2||0|| || || |
|T1||0|| || || |
|T2||0|| || || |
|T3||0|| || || |
|T4||0|| || || |
|S2||0|| || || |
* Vert. = Intended vertebral body
† No. = Number of misidentifications involving the intended vertebral body
‡ Errors Caudad = Wrong level is closer to the coccyx
§ Errors Cephalad = Wrong level is closer to the head
The importance of specifying the exact location of a spinal procedure on documents used for verification is illustrated by the following report:
The patient was brought into OR. Left side was verified with the physician. The physician announced Left 4. Surgical consent stated lumbar epidural injection with no level or side specified. Left 4 lumbar epidural injection was completed. The physician returned in the afternoon to perform other procedures and realized he had an incorrect schedule from his office, which led to the realization that he had done the wrong level on the morning patient.
The value of following the North American Spine Society (NASS) suggestion of using an intraoperative imaging study—after surgical exposure of the operative site, using markers that do not move—to confirm the vertebral level to be operated4 upon is illustrated by the following report:
Spinal procedure completed at unintended level. Discectomy was planned for L2-3 but performed at L3-4. This error was detected intraoperatively, and the surgery was extended to include the intended procedure. . . . Surgical site verification was performed according to our policy and did not include intraoperative imaging to confirm the level.
The importance of using stable markers to verify, rather than estimate, the intended vertebra is illustrated by this report:
An intraoperative fluoroscopy was used to localize the incision over C5-6. Fluoroscopy revealed that needle was at C4-5, so dissection was carried further distally to the presumed C5-6 level. The procedure was completed. However, a follow-up x-ray revealed that the site was C6-7.
The value of the NASS suggestion of an additional reading of the verification image by a radiologist, as well as the surgeon,4 is illustrated by another report:
The patient has a transitional lumbar vertebra. The surgeon performed surgery at the L3-4. The patient remained symptomatic. Review of the films with the radiologist determined that the radiologist and the surgeon were counting differently (one included the transition vertebra and one did not), resulting in confusion of the level for surgery. The patient was taken back to the OR, and surgery at L2-3 [was done] with good results.
Reasons for misreading the landmark vertebra included transitional vertebrae, lumbarization of the sacrum, prior fusions, osteoporosis, kyphosis, and obesity, among many other pathologies that could lead to confusion.
Other Wrong-Location Events
There were 59 intended procedures done on the correct patient but initiated at the wrong location other than the wrong side or wrong level. They constituted 11.8% of all 500 wrong-site surgery reports and represented 13.7% of all intended procedures done on the correct patient but initiated at wrong location on the body among the 430 patients for whom the location could be determined.
These procedures were initiated at locations near the correct locations:
- 14 procedures that started on the wrong part of the correct hand or fingers
- 11 procedures involving the removal of the wrong skin or subcutaneous lesion
- 9 procedures that started on the wrong part of the correct breast
- 7 procedures that started on the wrong part of the correct foot or toes
- 3 procedures that started on the wrong side of the correct elbow
- 3 vascular access procedures that started at other than the intended location
- 2 device implantations that started at other than the intended location
Unique events among the 500 reports involved the wrong tooth, the wrong cardiac vessel, the wrong part of the correct lung, the wrong midline hernia, the wrong part of the perineum, the wrong end of the femur, the wrong side of the internal knee, the wrong muscle, and an incorrectly placed incision over the lower spine. One event was described only as medial instead of lateral.
These events illustrate the importance of the following:
- Locations of procedures should be documented as specifically as possible (e.g., medial or lateral, proximal or distal).
- Marks referencing the location of the surgery on hands, fingers, feet, and toes should be placed as close as possible to the intended incision site.
- The exact location of skin and subcutaneous lesions should be marked.
Localizing breast lesions is more complex and may be discussed in a future issue of the Advisory.
Analysis of 500 consecutive wrong-site surgery reports identified the following themes:
- The process of having the patient state two identifiers to verify their identity appears to be effective in preventing wrong-patient errors.
- Preoperative documentation of the site of surgery should be specific enough for all OR team members to anticipate the correct location of the mark.
- Marks should be made as close to the intended incisions as possible. The exact location of skin and subcutaneous lesions should be marked.
- The most likely wrong-site error, by far, is a wrong-side error. Bilateral structures, especially extremities and eyes, are most likely to experience wrong-side surgery. The most common wrong-side error is the anesthetic block, accounting for 34% of all wrong-side errors and 21% of all wrong-site errors in the OR area.
- Some wrong procedures may result from surgeons becoming distracted during the operation. The OR team should maintain situational awareness of the intended procedures throughout the case, not just at the start of the case.
- Wrong-level spinal procedures represent 13% of all wrong-site procedures. Intraoperative misperceptions were reported nine times as often as errors based on misunderstandings of information available preoperatively. The prevention of wrong-level spinal procedures requires intraoperative verification of the correct spinal level.
- Mehtsun WT, Ibrahim AM, Diener-West M, et al. Surgical never events in the United States [online]. Surgery 2012 Dec 17 [cited 2013 Jan 29; epub ahead of print]. http://www.sciencedirect.com/science/article/pii/S003960601200623X.
- Ring DC, Herndon JH, Meyer GS. Case records of the Massachusetts General Hospital: case 34-2010: a 65-year-old woman with an incorrect operation on the left hand. N Engl J Med 2010 Nov;363(20):1950-7.
- Quarterly update on the preventing wrong-site surgery project: digging deeper. Pa Patient Saf Advis [online] 2010 Mar [cited 2013 Jan 29]. http://www.patientsafetyauthority.org/ADVISORIES/AdvisoryLibrary/2010/Mar7(1)/Pages/26.aspx.
- North American Spine Society. Sign, mark & x-ray (SMaX): a checklist for safety [online]. 2001 [cited 2013 Jan 29]. http://www.spine.org/Documents/SMaXchecklist.pdf.