Bone cement implantation syndrome is a rare but severe peri-operative pathology that presents on cementation of orthopaedic implants and has effects ranging from transient hypotension to cardio-respiratory arrest. Several reports have attempted to study its pathophysiology and establish risk factors that would allow identification of high risk patients and the undertaking of relevant precautions.
Bone cement is a chemical polymer, comprising poly-methyl-methacrylate (PMMA). It was introduced in the 1940s in clinical practice and was initially used to fix dental implants. It was first utilised in orthopaedics during the early 1960s by Sir John Charnley.1 In the case of hip arthroplasty, cementation techniques have since evolved from first generation (no cement plug, finger packing of femur) to fourth generation (vacuum mixing, pressurising and stem centraliser) techniques.
One of the issues that were soon identified with the introduction of cement in joint arthroplasty was a higher incidence of perioperative transient hypotension, pulmonary hypertension, hypoxaemia, cardiac dysrhythmias, metabolic acidosis and cardiac arrest, especially in elderly patients with cardiovascular comorbidities.1-4 These perioperative cardiovascular adverse events have been termed bone cement implantation syndrome (BCIS).
Classification systems have been proposed based on the severity of hypoxemia and hypotension.4,5 Several theories have since been suggested to explain this phenomenon and several relevant risk factors have been proposed. We aim to present these theories and risk factors in our paper.
BCIS is overall a rare event. It is, however, a clearly life-threatening event and one that can be anticipated and predicted to certain extend. At the same time, we feel that this pathology is grossly under-reported, mainly due to lack of recognition and lack of appropriate coding and death certificate completion.
We carried out a retrospective study in our institution in order to identify the mortality rate of cemented hemiarthroplasty involving the femur. We attempt to identify factors that put patients at higher risk of this syndrome.
We performed a literature search on PubMed and two electronic databases, namely EMBASE and Ovid Medline. We used search terms including ‘BCIS’, ‘bone cement syndrome’, ‘bone cement’ and ‘PMMA’. We yielded a number of studies and selected those that principally examined BCIS in hip arthroplasty.6,7 We also reviewed the published 2009 NPSA alert that examined the cement-related risks of hip arthroplasty,8 as well as the 2015 safety guidelines from the association of anaesthetists for Great Britain and Ireland, British Orthopaedic Association and the British Geriatric Society.9
Our institution utilises cemented and uncemented arthroplasties for the treatment of hip fractures and osteoarthritis of the hip and knee. We collected data retrospectively through the use of our incident reporting system and our theatre logbook. We identified the number of patients that had serious intra-operative morbidity or died during instrumentation and cementation of their femora from 2011 until 2015. After retrieving their complete medical records, we examined their pre-existing morbidities and attempted to detect risk factors for BCIS. Their notes were scrutinised, from admission to hospital up to their death and the subsequent issuing of their death certificate, with input from the coroner.
We finally reviewed their pre-operative radiographs in order to assess whether these patients would be suitable for an uncemented implant, based on the classification of proximal femoral anatomy as described by Dorr.10 We propose a protocol in managing patients with intracapsular femoral neck fractures that is already in use in our institution, aiming to prevent or reduce the incidence of BCIS in the future.
We identified six patients that passed away at the time of surgery or soon after, where there was documentation of adverse events starting from the time of cementation. Four of these patients passed away in theatre, whereas two passed away after their transfer to the intensive care unit. In all six cases, there was significant hypotension observed at the time of cementation and/or reduction of the implant, associated with a cascade of events that led to multi-organ failure and death.
In order to adhere to the strict criteria set to this study, we excluded three of these patients, as there was no explicit report of BCIS as the cause of death. We included the three patients in which the above was clearly documented after input from the coroner. These three patients included two cases of intracapsular hip fractures treated with a cemented hemiarthroplasty and once case of metastatic cancer to the distal femur, treated with a distal femoral replacement. All three cases are presented in the following table. (Table I).
All patients were assessed before surgery by an orthogeriatrician as per hospital protocol and optimised for theatre as much as possible. The two patients with hip fractures were operated on within 36 hours of admission to hospital.
As two of the patients that passed away were patients with intracapsular hip fractures, we decided to identify the total number of patients that had a cemented hip hemiarthroplasty between 2011 and 2015 and calculate the relevant mortality rate from BCIS. The total number was 296, giving a mortality rate of 0.68%. However, if we were to include all five cases of hip cemented hemiarthroplasty that were found to be very probable BCIS events, this rate would rise to 1.7%. We feel that therein lies one of the important issues with literature relevant to BCIS, in that this is an event that is under-reported either due to lack of recognition or lack of reporting tools that can be later used in retrospective studies or audits. Especially owing to the fact that the average in hospital mortality in this group is reported as high as 8%.11 There were no peri-operative deaths in the cohort of patients that received an uncemented hip hemiarthroplasty between 2011 and 2015.
Risk of death
According to a report issued by the National Patient Safety (NPSA) on the surgical risks of cemented arthroplasty for fractures of the proximal femur, there were 24 deaths and six cases of severe harm that were reported to the Reporting and Learning System (RLS) up to October 2008, with a potential link to cement insertion.8 These incidents were perioperative and were recorded as happening around cement insertion, although more details were not available, such as cementation technique and type and make of implants.
The NPSA reports that due to the voluntary nature of reporting such incidents to the RLS, the true prevalence of these incidents could not be established and the presented figures are most likely an underestimate of the problem.
The same NPSA report mentions that the Medicines and Healthcare Products Regulatory Agency (MHRA) database includes 20 deaths and four cases of severe harm relevant to cement use in hip arthroplasty, between 2000 and 2008. These cases are not duplicates of the ones reported in the RLS according to the NPSA.
Combining data from the MHRA and the NPSA, there are a total of 44 deaths and 10 cases of severe harm relevant to cement up to 2008 in England. In response to the aforementioned NPSA safety alert, Hossain et al12 reported on their five-year mortality in patient undergoing cemented and uncemented arthroplasties for hip fractures and suggest that there was a significantly higher risk of death in cemented cases. All of their deaths were in patients with severe cardiopulmonary comorbidities. This is clearly mirrored in our study.
In 2014, Rutter et al13 researched the risk of death relevant to cemented hip hemiarthroplasty by examining National Reporting and Learning system (NRLS) data. They suggested that there were 62 deaths reported between 2005 and 2012, directly relevant to cement implantation. They reported that there was one death for every 2,900 cemented hemiarthroplasties for that period of time, a risk of 0.034%. They suggested that this figure is most likely an underestimation of the true prevalence of this pathology, due to under-reporting.
In terms of further comparing mortality rates between cemented and uncemented method of fixation of hemiarthroplasty implants, a systematic review that was published by Ahn et al14 in 2008 did not demonstrate any increase in peri-operative and long-term mortality between the two groups. A meta-analysis by Li et al,15 a study presenting data from the Australian joint registry,16 a randomised-controlled trial by Taylor et al17 and another randomised-controlled trial by DeAngelis et al18 have all reached the same conclusion. A more recent study, which again was a randomised-controlled trial, was published in 2015 by Inngul et al.19 They report no statistically significant difference in the mortality rates between the two methods, using contemporary implants.
A closer look at this study, however, demonstrates that it was not powered to look for differences in mortality specifically and that the authors pooled cases of hemiarthroplasties and total hip replacements, as recruitment was slower than anticipated.
Finally, although no statistical difference existed between the two methods, there were more deaths in the cemented group at four months after surgery, which again raises the issue of powering such a study for a rare event like BCIS.
Prashanth et al20 also report no difference in outcome during a comparative study of bipolar implants. A valid point that Rutter et al raised as criticism of these studies, is that they had relatively small sample sizes and were unlikely to pick up a clinically significant difference in mortality rates between cemented and uncemented implants.13
A large retrospective study from Parvizi et al21 reported 23 intra-operative deaths in a total of 38,488 hip arthroplasty cases, all of them in cases where cement was used. There was a higher reported risk of death in hemiarthroplasties compared to THAs and in patients with previous cardiovascular comorbidities. In the majority of patients who died and underwent a subsequent autopsy, bone marrow microemboli were found in the lungs.
Interestingly, they report that in the years following this study, measures taken in order to reduce femoral canal pressures during cementation have greatly reduced the mortality rates on insertion of cement. In our study, all patients that passed away following cemented hip hemiarthroplasty had Dorr10 type B proximal femora, therefore permitting the use of uncemented implants that would potentially avoid BCIS.
A prospective study from Li et al23 reported on the findings of invasive monitoring of blood pressure and oxygen arterial pressures in patients undergoing cemented THAs and hemiarthroplasties with third generation cementing techniques. No adverse changes or deaths were detected in that study.
There is now evidence that preparing the femoral canal with a brush and pulsatile lavage and drying prior to instrumentation of the femur significantly reduces the incidence of fat embolism.7,6,24 Vacuum mixing of cement seems to lead to fewer embolic episodes when compared with conventional mixing.25 The use of a cement restrictor in combination with retrograde gun cementation has been suggested to cause a less pronounced physiological disturbance when compared to other methods of cementation.26 Studies on animal models have demonstrated that uncemented implants carry a minimal risk of cardiovascular disturbance during insertion when compared to cemented implants.27
A recent study presenting early and late mortality rates in elderly patients undergoing hip hemiarthroplasties supports the use of uncemented implants in these patients over cemented ones, although the rate of re-operation was higher in the uncemented group.28 Of note also from an anaesthetic perspective, single epidural anesthesia has been shown to have a higher risk compared to combined spinal epidural.29
When attempting to identify patients at risk of severe cardiovascular cement-related complications, a common finding is that patients with pre-existing cardiovascular compromise and ASA grades of 3 or 4 are at higher risk and should be monitored very closely.5,30
A prospective study from Leidinger et al identified patients with pre-operative pulmonary hypertension as a high risk group for BCIS31 and Olsen et al32 consider patients on diuretics, warfarin and those who suffer from chronic obstructive pulmonary disease as high risk patients. Another prospective study identified patients undergoing reaming for pathological fractures as high risk for embolic events, with a mortality rate reported as high as 4.3%.33 In any case, optimised peri-operative hydration is key in reducing the risk of severe events.24
A recent systematic review by Rocos et al suggests that under resuscitation is a very common problem and this requires further research, with pre-operative hypoperfusion and anaemia being the main issues. The three patients that passed away in our study presented with significant cardiovascular comorbidities and in two out of three cases, with diagnosed malignancy.
When comparing mortality rates in groups with different indications, patients with intracapsular fractures have a rate of 0.2% and in patients with pathological fractures the mortality rate significantly rises to 4.3%.24 Implantation of long-stemmed femoral components increases the risk of BCIS.34 Patients with previously un-instrumented femora are at a higher risk of BCIS when compared to patients who undergo revision surgery.5
The majority of embolic events and cardiovascular changes have been detected during the stages of reaming the femur, cementing, inserting the stem and reducing the joint.33,35 Several mechanisms have been suggested to explain the risk of massive embolism following instrumentation and cementation of the femoral canal in hip arthroplasty cases, with the most widely acceptable being that of the embolic or pressure-model.5
The common endpoint in BCIS is the intravasation of fat, air, cement particles, aggregates of platelets and fibrin and bone marrow5 in the metaphyseal femoral veins and the venous system located along the linea aspera. These particles embolise to the lungs leading to right ventricular dysfunction, or even paradoxically to the cerebral and systemic circulation, through either a patent foramen ovale or through transit from the pulmonary circulation. Severe osteoporosis has been suggested as resulting in reduced medullary fat support36 and increased marrow venous migration following reaming and cementation. Similarly, it has been suggested that reaming the femoral canal prior to cement and stem insertion, disrupts the medullary veins, creating a direct shunt through which bone marrow and fat can enter the circulation.33 Furthermore, it has been demonstrated that reaming and cementing may result in pressures as high as 800mmHg in the distal femur,37 whereas insertion of the prosthesis can lead to pressures of the order of 1400mmHg.30 The fact that PMMA does not directly cause cardiovascular changes in patients, despite high levels of PMMA having been detected in venous blood after cementation,33 is supported by an animal model where the same changes were demonstrated whether cement or bone wax was inserted and pressurised in canine femora.27
A prospective study identified patients undergoing reaming for pathological fractures as high risk for embolic events
Using conventional cementation, a randomised controlled trial by Pitto et al38 demonstrated that during the insertion of the cement and the stem severe embolic effects were evident in 85% of patients, arterial oxygen saturation significantly decreased and pulmonary circulation pressures increased. Apart from the latter effect, these changes were not evident when uncemented implants were used and when cemented stems were introduced with the use of a continuous vacuum system with two cannulae introduced in the femur.30 One of the issues with this technique is that the vacuum system is not easy to reproduce as it involves the insertion of proximal and distal drainage cannulae in the femur.
The same author published the results of a randomised controlled trial in 2002 where they used a single drainage cannula inserted in the linea aspera and connected to a negative pressure system. Patients that had this system had significantly less incidence of deep vein thrombosis and reduce incidence of intra-operative embolic events.39 Although not all centres use proximal femoral cannulae during cement and implant insertion, it is now common practice to insert a flexible suction catheter in the femur that remains in place while cement is introduced, but this has to be removed as the stem is inserted in the cement mantle.
Cardiovascular changes during femoral cementation and implant insertion seem to be very frequently, if not always, present.4There is a real need to identify high-risk patients, whose physiology cannot tolerate this significant insult and will result to severe cardiovascular failure. We suspect that the previously published incidence of BCIS in several large series underestimates the issue, with the principal cause being that of inadequate recognition and therefore insufficient documentation and coding. Based on the available literature, we propose that patients who are at high risk of BCIS should be actively identified pre-operatively. Surgeons could consider an uncemented implant, which carries the additional benefit of reduced operative time,15 respecting the higher risk of intra-operative femoral fracture in using these implants15 and the fact that some studies have not identified a reduced peri- or post-operative mortality with the use of these implants.16,40 Patients should be optimally hydrated peri-operatively and modern cementation techniques, including the use of flexible suction catheters, should be employed as per national guidance.9 Another useful protective step would be the use of low-viscosity cement that does not seem to compromise the pullout strength significantly.41
Insertion of a negative pressure suction system in the proximal femur could prevent embolic events and could be considered in high-risk patients. Intra-operative steps to identify this pathology could include the use of trans-oesophageal echocardiography that would demonstrate major embolic events in real time during cement and stem insertion. Combined spinal epidural anaesthesia can also be considered. Finally, vigilance on behalf of the anaesthetist will allow early identification of these life-threatening events and the institution of early measures such as vasopressors and invasive monitoring. On account of all published literature and of findings in our local unit, we have implemented a protocol for the identification of high-risk patients and steps to prevent the incidence of BCIS, which is in addition to the national guidelines (Box 1).
Angelos Assiotis, Specialist Registrar, Trauma and Orthopaedics, Hillingdon Hospital
David Ahearne, Consultant, Trauma and Orthopaedics, Hillingdon Hospital
Ioannis Pengas, Consultant, Trauma and Orthopaedics Basingstoke and North Hampshire Hospital
Conflict of interest: This study was carried out at the Hillingdon Hospital, London, UK. No benefits in any form have been received by any one of the authors, relevant to the subject of this article. All authors declare no conflict of interest. The Hospital Ethics committee gave ethical approval for this study.
1. Urist MR. Acrylic cement stabilized joint replacements. Curr Probl Surg 1975;:1–54 http//www.ncbi.nlm.nih.gov/pubmed/803182 (accessed 31 Dec 2017)
2. Borroff M, Green M, Gregg P, et al. NJR ANNUAL REPORT 11TH. www.njrcentre.org.uk (accessed 31 Dec 2017)
8. NPSA. Patient safety resources. http://www.nrls.npsa.nhs.uk/resources/?EntryId45=59867 (accessed 31 Dec 2018)
13. Rutter PD, Panesar SS, Darzi A, et al. What is the risk of death or severe harm due to bone cement implantation syndrome among patients undergoing hip hemiarthroplasty for fractured neck of femur? A patient safety surveillance study. BMJ Open 2014; 4
14. Ahn J, Man L-X, Park S, et al. Systematic review of cemented and uncemented hemiarthroplasty outcomes for femoral neck fractures. Clin Orthop Relat Res 2008; 466: 2513–18 doi:10.1007/s11999-008-0368-3
19. Inngul C, Blomfeldt R, Ponzer S, et al. Cemented versus uncemented arthroplasty in patients with a displaced fracture of the femoral neck: a randomised controlled trial. Bone Joint J 2015; 97–B: 1475–80
27. Orsini EC, Byrick RJ, Mullen JB, et al. Cardiopulmonary function and pulmonary microemboli during arthroplasty using cemented or non-cemented components. The role of intramedullary pressure. J Bone Joint Surg Am 1987; 69: 822–32
31. Leidinger W, Hoffmann G, Meierhofer JN, et al. [Reduction of severe cardiac complications during implantation of cemented total hip endoprostheses in femoral neck fractures]. Unfallchirurg 2002; 105: 675–9
32. Olsen F, Kotyra M, Houltz E, et al. Bone cement implantation syndrome in cemented hemiarthroplasty for femoral neck fracture: incidence, risk factors, and effect on outcome. Br J Anaesth 2014; 113: 800–6
38. Pitto RP, Koessler M, Kuehle JW. Comparison of fixation of the femoral component without cement and fixation with use of a bone-vacuum cementing technique for the prevention of fat embolism during total hip arthroplasty. A prospective, randomized clinical trial. J Bone Joint Surg Am 1999; 81: 831–43
39. Pitto RP, Hamer H, Fabiani R, et al. Prophylaxis against fat and bone-marrow embolism during total hip arthroplasty reduces the incidence of postoperative deep-vein thrombosis: a controlled, randomized clinical trial. J Bone Joint Surg Am 2002; 84–A: 39–48
40. Bell KR, Clement ND, Jenkins PJ, et al. A comparison of the use of uncemented hydroxyapatite-coated bipolar and cemented femoral stems in the treatment of femoral neck fractures: a case-control study. Bone Joint J 2014; 96–B: 299–305