Patient Safety Tip of the Week

November 16, 2010

Lost Lab Specimens



In our September 16, 2007 Patient Safety Tip of the Week “Errors in the Laboratory“ we discussed the tragic issue of pathology specimens being mixed up. In such cases a patient may undergo an unnecessary surgical procedure for an erroneously diagnosed cancer, and the other patient may have a delayed diagnosis of his cancer.


Another potentially tragic error is loss of surgical specimens. How would you feel if you underwent an invasive procedure with biopsy and your specimen got lost? At best, you could undergo another biopsy, though you’d have to endure the risks and inconveniences of that second procedure. At worst, there would be no place to look for more tissue (eg. after removal of a skin lesion suspected of being a malignant melanoma) and your anxiety levels would soar over the uncertainty.


There aren’t any great statistics on how often surgical specimens are lost. One study (Makary 2007) looking at surgery in an outpatient clinic or a hospital operating room found 4.3 surgical specimen identification errors in every 1000 cases (and about 60% of those errors were in biopsies) but did not note cases of lost specimens.  The Pennsylvania Patient Safety Authority did a Patient Safety Advisory “Lost Surgical Specimens, Lost Opportunities” in 2005. They had received at least 30 reports of lost surgical specimens through their patient safety reporting system but did not provide any denominator data so the actual rates of lost specimens is unknown. Fortunately, these are probably rare events. However, they can be devastating to patients and staff when they occur.


After a near-miss where surgical biopsy specimens were temporarily misplaced (but fortunately found in a few days), we helped in a root cause analysis (RCA) to get a better understanding of the vulnerabilities. Technically, it was really a FMEA (Failure Mode and Effects Analysis) since we went well beyond the RCA and looked at the entire pathway that a specimen would follow from the OR (or office) through final reporting of pathology.


We began by flowcharting the entire process. After you do your initial pass at flowcharting, you need to go back and ask questions like “How is that handled after hours (at night or on weekends)?”.


Key to doing a good RCA or FMEA is involving the key players who are closest to the process. They are the ones most likely to tell you where the problematic points are (and tell you the solutions!). You also need to visually observe the entire process as it plays out in real time. You will be surprised at how often you see things that others have come to accept as “normal”. And you’ll see the various “workarounds” that have crept into the process.


Many who have done similar RCA’s or FMEA’s have chosen the start of the process to be when the specimen is taken from the patient in the OR. However, it really should begin well before that. When a case is booked onto the OR schedule one of the questions that should be asked is “Do you anticipate there will likely be surgical specimens to be sent to the lab?”. Then, during the “pre-op huddle” or “pre-op briefing” that question should be asked again.


The OR surgery/procedure documentation sheet should contain a section detailing what specimens were obtained and sent to the laboratory. In fact, we recommend you also have a check box for when no specimen is sent and have the surgeon sign or initial that box when no specimens are sent.


Using forms or checklists may help remind people what to do with the specimen. If you use the World Health Organization’s  Surgical Safety Checklist the Sign Out phase is completed before the patient leaves the operating room. The Checklist coordinator verbally confirms the name of the procedure, the instrument and sponge counts, correct labeling of any specimens, any equipment problems, and key concerns for postoperative care and recovery.



Labeling specimens should take place immediately when the specimen is removed in the OR. Specimens are usually given to a nurse in the OR, who then puts the specimens in a container with the patient’s identification (at least 2 methods of patient identification), date/time taken, surgeon’s name, and description of the tissue (what it is and where it was taken from) and a requisition slip with similar information. The containers with the specimens are then generally placed in a receptacle in the “pick-up” area of the OR nurses’ station and entered into a log book. Specimens are in formalin in the small containers but larger specimens need to go in larger containers. Formerly, nurses in the OR would add formalin to the larger containers but that practice had been stopped for safety reasons (so formalin is added to the larger specimens in the lab itself). The specimens are then taken by someone (in this case, that “someone” could be a nurse, a courier, or anybody who happened to be going to the lab). That person would sign the log book in the OR and take the specimens to the lab. At the lab they would sign a log book, as would the person receiving the specimens in the lab. After logged in at the lab the specimens are taken by the technician to the “grossing” section and the technician prepares cassettes to be matched to each specimen. The pathologist then does a gross examination (dictating his/her report) and places specimen parts into the labeled cassettes. From there the specimens are put into the “processing” machine overnight. A technician maintains the lab log and indicates on there how many cassettes the pathologist listed in his dictated report. However, sometimes the technician does not hear the dictation until the following morning. In the morning the technician removes the cassettes from the processing machine and takes them to the “embedding” machine (paraffin bath). From there they go to a microtome where slides are prepared and labeled. They then go to a warming oven to dry, then are cooled off, and stained. Once stained and dried the slides are placed in a slide holder and placed, along with the corresponding gross descriptions, on the pathologist’s desk. The pathologist adds the microscopic examination to his report and the slides are put in a filing cabinet for storage. This is the process that takes place at most labs.


Several similar RCA’s or FMEA’s using this type of flowcharting appear in the literature (Slavin 2001, Messinger 2010). One good article on issues related to lost specimens (OR Manager 2007) provides some examples that occurred at hospitals. It provides details about policy and procedure changes that were implemented after a lost specimen, including a copy of the specimen handling policy adopted at that hospital (Swedish Medical Center 2007).


Chain of custody is a term usually associated with precise tracking of forensic specimens or evidence in legal cases. Most labs at times do have to adhere to strict chain of custody procedures when they do handle items involved in litigation. Yet shouldn’t every specimen be handled with the same level of scrutiny and security? I would certainly hope any specimen of mine were handled that way! Chain of custody is nicely described in the PPSA Patient Safety Advisory “Lost Surgical Specimens, Lost Opportunities” and the OR Manager article. It usually involves logs containing the names/departments releasing the specimen, names/departments accepting the specimen, dates and times of such transfers, patient identification, specimen number, specimen description, and others.


Lacking are good studies demonstrating the impact of new technologies (barcoding, RFID’s, etc.) on tracking lab specimens. We don’t doubt that these will play a primary role in specimen tracking very soon. After all, if Fedex and UPS can tell you where your package is at any moment in time, why can’t a hospital do the same? Two articles (Malone 2010, Soom 2010) in the July 2010 issue of Clinical Laboratory News discuss those new technologies and their role in tracking specimens.


One very simple intervention that should be done at all handoffs is a specimen count. If you knew you had 14 specimens or cassettes at Step B and you now have only 13 at Step D, you have at least isolated the locations where the specimen was most likely lost or misplaced. Its just like doing a sponge count when your are doing surgery. (Speaking of sponges…one cause of “lost” specimens is when tiny tissue specimens get attached to small sponges that may be put in specimen containers).


During the discussion with key stakeholders, anecdotal items tend to come out. Apparently sometimes the receiving person in the lab after hours did not always put the larger specimens in formalin. Pathologists often grumble that the surgeons don’t provide enough detail about the specimens.


What happens after hours probably differs from facility to facility. In our RCA the OR nurse would deliver the specimen to the lab and turn it over to whoever was working in the lab that night (that might be someone other than the normal tissue lab technician). One must be certain that the person receiving specimens after hours is as fully trained in proper handling and processing of those specimens as your day staff are.


Specimens coming from outside the hospital via courier (eg. from offices) can be especially problematic. You may have little control over the labeling and handling that occur in those offices. And there are lots of anecodotes of such specimens never reaching the lab. It may make sense to have some communication from the offices to the lab ahead of time, notifying them that a specimen or specimens are on their way to the lab. In any event, the logbook and chain of custody procedures should take place every time a handoff takes place (eg. office-to-courier, courier-to-lab, etc.) and specimen counts should match.


One other caveat that comes from the literature on mislabeling of specimens: if you can avoid it, don’t batch! Deal with specimens one at a time. Any time you are dealing with labeling multiple specimens at the same time you run the risk of transposing labels.


Ownership” becomes a key element. In the Slavin article, the pathology department decided that it would take ownership of the transport process from the OR to the lab. That effectively reduced the number of handoffs. Handoffs are always opportunities for errors to occur so anything that reduces the number of handoffs generally improves safety. In our own RCA/FMEA the OR staff decided to take ownership for the process when done after hours (taking the specimen to the lab, entering it into the log there, and putting it in formalin). This step basically says “we will not assume that someone else will do it”. Another example was our lab staff deciding to take over removal of the waste receptacles (rather than housekeeping) so they would have control in the event of a missing specimen.


Having a checklist “where to look for lost specimens” may also help. This should include places where previously misplaced specimens were ultimately found and should point staff to review the chain of custody logs to identify where the specimen was last recorded.


Having “outsiders” (i.e. people who do not ordinarily work in the pathology lab) observe the process also provides an opportunity to observe the work environment. That is where the little things, that escape the attention of the lab personnel, may be identified. For example, you might notice that the trash receptacles sit in locations where it might be possible to accidentally knock a specimen or cassette into them (don’t be afraid to ask them “Have you ever knocked a cassette into that trash can?”). Or you might notice that all the specimen containers and their matching cassettes are lined up like dominoes and vulnerable to being knocked over by an errant elbow!


Review of the work environment should include attention not only items like lighting and clutter but also workload issues. Mistakes tend to occur most often under stressful conditions, such as time pressures or deadlines, understaffing, fatigue, conflicting goals and incentives, etc.


Any time we implement a quality improvement project we need some metrics to track and demonstrate that our changes induced positive outcomes. The ultimate outcome goal is obviously to avoid any lost specimens at all. Given the rarity of lost specimens, monthly tracking of “lost specimens” is likely to produce lots of useless graphs. Hence, some organizations looked for proxies or surrogate measurements that could be used to demonstrate improvement. One (Messinger 2010) chose to monitor “misplaced” specimens (where specimens were temporarily lost but eventually found) and was able to use control charts to monitor processes. Another (Slavin 2001) used timeliness of specimen processing as a proxy for potential lost specimens. They felt that since overall processing time is a function of how long a specimen sits in one or more areas, this would be a good surrogate for potential lost specimens. They incorporated that measure into control charts and were able to demonstrate that their QI project was producing desirable results.


Education, training and competency assessment are, of course, essential to quality in any laboratory setting. But don’t forget you also need to make sure everyone who might be in that laboratory, including your housekeeping staff, understand security of specimens.


Prompt communication upon identifying a lost specimen is critical. That communication starts the search for the missing specimen. Such cannot be delayed. For instance, if the specimen is in a trash can it may be gone and irretrievable by morning. Similarly, if your specimen count on arrival at the lab does not match what was in the logbook at the OR end, an immediate call to the OR is needed to avoid having a specimen discarded along with the other surgical paraphernalia.



Obviously, this is a great topic for your organization to consider for a FMEA (Failure Mode and Effects Analysis). Flowcharting the entire process from pre-procedure through the entire pathology process will help you identify not only potentials for losing specimens but also those potentials for mislabeling specimens and mixing them up.





Update on specimen mixups: Since our September 16, 2007 Patient Safety Tip of the Week “Errors in the Laboratory“ where we discussed pathology specimens being mixed up, we have come across some other good articles on mislabeling and mixups of surgical specimens (Layfield 2010, Lippi 2009 , Nakhleh 2003, Valenstein 2004, Astion 2006).



Update on lost lab specimens: See our March 20, 2018 “Minnesota Highlights Lost Tissue Samples







Makary MA, Epstein J, Pronovost PJ et al. Surgical specimen identification errors: A new measure of quality in surgical care. Surgery 2007; 141(4): 450-455



Pennsylvania Patient Safety Authority. Lost Surgical Specimens, Lost Opportunities.

PA PSRS Patient Saf Advis 2005 Sep; 2(3):1-5



WHO Surgical Safety Checklist



Slavin L. Best MA. Aron DC. Gone but not forgotten: the search for the lost surgical specimens: application of quality improvement techniques in reducing medical error.

Quality Management in Health Care 2001. 10(1): 45-53



Messinger B, Ahlin P. Patient Safety Focus: Lost Versus Misplaced Samples. Improving Lab Quality with Pull Measures. Clinical Laboratory News 2010; 36(1): January 2010



OR Manager. Taking steps to protect patients from specimen-handling errors. OR Manager 2008; 24(12): 1, 7-8



Swedish Medical Center. Specimens: Labeling and Handling (Clinical Procedure). August 2007. Downloaded from OR Manager Toolbox.on website.



Malone B. Radio Frequency Identification of Specimens. Can it Solve Labs’ Preanalytical Predicament? Clinical Laboratory News 2010; 36(7): July 2010



Soom S, Strathmann F. Patient Safety Focus: How Labs Can Protect Against Specimen Loss . Clinical Laboratory News 2010; 36(7): July 2010



Layfield LJ. Anderson GM. Specimen labeling errors in surgical pathology: an 18-month experience. American Journal of Clinical Pathology. 134(3):466-70, 2010 Sep.



Layfield LJ. Anderson GM. Specimen labeling errors in surgical pathology: an 18-month experience. American Journal of Clinical Pathology. 134(3):466-70, 2010 Sep.


Lippi G, Blanckaert N, Bonini P, et al. Causes, consequences, detection, and prevention of identification errors in laboratory diagnostics. (Review). Clinical Chemistry & Laboratory Medicine 2009;  47(2): 143-153, February 2009.;jsessionid=MgDHHcqjQPwY1QrtnFLQwKF25z3m17Lf2vQkC6mjpLkvQVQ7RNtm!1173397540!181195629!8091!-1



Valenstein PN, Sirota RL. Identification errors in pathology and laboratory medicine. Clinics in Laboratory Medicine 2007; 24(4): 979-996



Astion M. AHRQ WebM&M. Right Patient, Wrong Sample. December 2006



Nakhleh RE. Lost, Mislabeled, and Unsuitable Surgical Pathology Specimens. Pathology Case Reviews 2003; 8(3): 98-102,_Mislabeled,_and_Unsuitable_Surgical.3.aspx















Patient Safety Tip of the Week Archive


What’s New in the Patient Safety World Archive