The Bone Cement Hypercoagulation Syndrome: Pathophysiology,Mortality,and Prevention

Abstract

Since Charnley introduced acrylic cement to seal metallic hip prostheses in the 1950s, reports of perioperative fatal cardiorespiratory and vascular dysfunctions have been published. Studies on humans and animals have shown neurogenic stimulation and substantial local and systemic activation of coagulation are caused by surgical bone marrow damage and chemical cell destruction by toxic monomeric methyl methacrylate from the implanted cement and other tissue-released substances. Venous blood-borne cell fragments and conjugates of activated cells from the surgical site are sequestered and trapped in the pulmonary microcirculation. A substantial hypercoagulation occurs in the lung circulation. Hypercoagulable blood is passed over to the arterial side and may cause vessel obliteration and organ damage. This process may affect the brain, heart, and kidneys and, through the release of vasoactive substances, introduce hemodynamic imbalances that can lead to fatal outcomes in susceptible populations such as elderly patients with hip fractures. The main underlying pathophysiologic processes leading to these occasionally devastating outcomes are a substantial activation of coagulation and cell destruction caused by the toxic substance released by curing bone cement and several vasoactive substances.

Introduction

Implantation of bone cement in patients with hip prostheses is associated with intraoperative cardiorespiratory and vascular instability. The syndrome has been explored for many years, and a great deal of knowledge has been accumulated. Despite this, little is understood about the pathophysiology of this potentially fatal condition.^1-6 Therefore, it seems pertinent to summarize the current status of this syndrome's understanding and the cascade of events that can lead to it, as shown in Figure 1.

Figure 1.

Bone cement hypercoagulability syndrome pathophysiologic cascade.

Following the First World War, acrylic resins were used by dentists as filling material. Later, surgeons composed prostheses from this plastic substance but without success. Around the Second World War, metallic hip prostheses were manufactured and installed, but even these failed. In the 1950s, a British surgeon, Charnley, used an unpolymerized acrylic cement to seal metallic hip prostheses. This technique succeeded and allowed the patients to walk without pain. However, the downsides were reports on fatal cardiorespiratory and vascular dysfunction intimately linked to the surgical implantation procedure.^7-17 We analyzed the historical literature to systematically examine the hypotheses, autopsies, and interdisciplinary pathophysiological studies carried out in vivo and vitro. Our aim was to elucidate the most likely triggers of the severe clinical events intimately linked to the surgical procedure.

Autopsy Findings

In the pre-prosthetic period, fat cells were commonly found in the organs of patients who had passed following treatment for large bone fractures. ¹⁸ Autopsies of major joint prostheses of patients who were operated on have also shown fat droplets, in addition to a spectrum of findings ranging from normal organ tissue architecture to pneumonia; air bubbles and emboli of cell conjugates in the vasculature of the lungs, brain, and kidneys; classical venous thromboembolism; bone cement-derived methacrylate particles in the lung vasculature; and myocardial infarction. Among those that died from respiratory distress, microemboli and fibrin deposits were commonly found in the small vessels of the lungs.^{13,15,16,18-23}

Bone Traumatization, Venous Microembolization, Thrombin, and Complement Effects

In prosthetic surgery, manipulation applied to the femoral bone cavity and implantation of cement and prosthesis causes intravasation of bone marrow cells that is similar to post-mortem findings of the pre-prosthetic era.^18,24,25 Conjugates of cells and lumps of bone marrow debris have been visualized on transesophageal Doppler as showers of embolic aggregates passing through the right side of the heart in cemented individuals but markedly less in uncemented patients.^26-30

From the venous side, large amounts of cells and thrombin are transferred to the arterial side, through vessel shunts in the lungs and patent foramen ovale in the heart with eventual widespread distribution to the body.^31-37

These aggregates contain activated cells which interact with tissue factor exposed on the luminal side of endothelium and on the unmasked subendothelium of microtears infiltrated with platelets, leukocytes, and erythrocytes. This phenomenon not only occurs at the surgical site but also in vessels at distant cites.^38-41 In animals, intravenous injections of adipose tissues containing small vessels with tissue factor or administration of thrombin caused accumulation of fibrin and platelets and a high number of deaths due to respiratory distress.^42,43 In patients with high levels of tissue factor-like components, accumulation of radiolabeled fibrin and platelets in the pulmonary vasculature has been noted during bone-shaft preparation and impaction of cement and prosthesis.^44,45 These accumulations have been measured as a greater than 200-fold increase in arterial fibrinopeptide-A levels which is comparable to massive thrombin activity which acts to convert non-thrombogenic endothelium into a prothrombotic surface which then amplifies the blood coagulation process, a cascade significantly augmented by impacted cement.^{34,35,38,40,41,46,47}

Thrombin is a pivotal enzyme in the coagulation process and a potent activator of platelets. It transforms fibrinogen to fibrin, of which 5% to 10% is deposed in the pulmonary vasculature during cemented hip substitute surgery. In the ensuing fibrin meshes, blood cells are trapped and eventually adhere to the activated endothelium on the exposed subendothelial matrix. This process causes mechanical obliteration of vessels.^34,35,45-48

Thrombin also has hormone-like effects, operating at varying thresholds. For example, while picomolar concentrations cause the production of nitric oxide in vascular endothelial cells, lower to upper nanomolar concentrations cause contractions of smooth muscle cells. This induces cell-shape changes of the endothelium which increase the permeability of albumin, the consequences of which are relatively increased in the vessels comprising the pulmonary vasculature as they increase the exposure of subendothelial procoagulant constituents.^39-41,48-51 This further promotes aggregation of platelets and granulocytes, thereby causing postcapillary vasoconstriction and subsequent increases in pulmonary arterial pressure.^52-54 Activated and destroyed cells release several substances such as serotonin, prostanoids, histones, and RNAs, all of which intensify hemodynamic alterations that persist chronically in some patients. A subset of the cell-released molecules disseminates systematically and acts as cytotoxins which cause organ damage both proximally and distally.^24,55-58 Overall, the intravascular coagulation process correlates with decreases in intraoperative peripheral blood pressure and detrimentally impacts the efficacy of gas exchange.^24,25,59

Coagulation and inflammation are linked through the binding of thrombomodulin with the endothelial cell surface and serine proteases.^60,61 Through crosstalk between proteolytic enzymes in the coagulation and complement systems, tissue factor is upregulated on circulating leukocytes and endothelial cells which favors cellular interactions; however, an overactivation of the complement system has been found to lead to tissue damage and induce deleterious effects.^61,62 During hip replacement, a moderate reduction of factor C4 and fragment C3(c) has been observed although without significant correlation with cementation.^34,63,64 However, 30 min to 24 h after impaction of cement and prosthesis implantation, reduced plasma whole complement activity (CH50) and increased levels of anaphylatoxins (C3a and C5a) have been found in hip surgery and after trauma, suggesting a delayed inflammatory response caused by massive intraoperative activation of coagulation.^62,65 An intimate feedback system between the coagulation and complement proteases favors a prolonged prothrombotic environment.^65,66 Thus, the delayed timing of anaphylatoxin formation following the critical phase of surgery excludes allergic reactions as the immediate cause of cardiorespiratory and vascular deterioration during cemented hip substitute surgery. However, the interplay between coagulation and complement may give momentum to ongoing processes and contribute to postponed severe adverse events. ⁶²

Plasma concentrations of vasoactive substances, for example, thrombin, monomethyl methacrylate (the liquid component of implanted cement), serotonin, thromboxane (B2), and prostaglandin F1α, all increase during surgery; however, interleukin-6 only rises post-operatively, suggesting it plays a role in later inflammatory-driven complications (vide supra).^34,65,66 In canines, inhibition of serotonin has been possible through heparinization prior to surgery; however, administration of methylprednisolone has not been found to modify a hemodynamic prostanoid response.^56,67 Histamine has also been implicated as a cause of cement-implanted cardiorespiratory and vascular reactions; however, no differences in plasma histamine levels were found when patients operated with cemented prostheses were compared to individuals with uncemented implants.^68,69 Although only a limited number of vasoactive substances have been investigated specifically in hip surgery, previous studies have strongly implicated the involvement of inflammatory and coagulative factors similar to other major surgeries.^65,70,71

Early during surgery, preparation and mechanical manipulation of the femoral shaft both trigger moderate autonomic reflexes while embolized cell clusters from the surgical site mediate vasoactive substances, the consequences of which contribute to pulmonary hypertension.^57,58,72,73

Bone Cement Released Methyl Methacrylate Monomer and Impact on the Coagulation System

Almost 70 years have passed since monomeric methyl methacrylate (MMA) was found to cause fatal reactions during hip replacement. In the early era of bone cement implantation, studies investigating MMA found that respiratory arrest was the primary cause of death rather than cardiac arrest.^74-76 Indeed, although the functions of both the cardiac and respiratory systems were depressed in a dose-dependent manner, the respiratory system was found to be more sensitive to MMA than the myocardium.^77-79 Through monomer infusion into the inferior vena cava in canines, 1250 µg/ml was determined as the dose at which respiratory arrest occurred while the heart continued to beat. ⁷⁶ Other studies investigating the effects of MMA in animal models found common adverse effects including soft tissue damage such as congestion, edema, and thrombosis.^80-82

Blood MMA levels in patients undergoing joint replacement vary with sampling site and collection time relative to cementation with mixed venous levels up to 200 mg/100 ml and peripheral vein levels up to 1.3 mg/100 ml.^83-85 Pure monomer enters venous blood at the site of administration and eventually undergoes a 10- to 100-fold reduction in concentration due to dilution and liposolubility before reaching the pulmonary vasculature. MMA concentrations found in mixed venous blood in animal studies are consistent with measured human blood concentrations. MMA rapidly penetrates lipid membranes of blood cells, endothelial coverings, and nerve sheets. MMA is partially removed from plasma through expired alveolar gas, and, by the time it enters arterial circulation, its concentration undergoes a subsequent 10-fold reduction in concentration. Pharmacodynamic studies found MMA levels in the pulmonary vasculature peaked within 1 min of femoral shaft cementation.^76,85-88 However, despite various studies documenting the extent of MMA cytotoxicity, some have claimed the amount of monomer as measured in human blood was not likely to have triggered fatalities.^{7,24,78,83,88–90} This claim was countered, with groups citing the pharmacodynamics of MMA leading to maximum concentrations easily passing unrecognized and accounting for the divergence in levels sampled from different sites at different time intervals following cementation.^{6,75,84,85,87,88} High concentrations of MMA in peripheral venous blood close to the cemented bone marrow cause cytolysis, release of proteolytic enzymes, and exposure of subendothelial procoagulant constituents thereby inducing coagulation.^91-97 Due to co-stimulation alongside thrombin, even moderate concentrations of MMA are able to trigger cell membranes to interact with plasma coagulation factors. ⁹⁷

MMA's effects on the vasculature tree lead to hemodynamic instability beyond the impairment caused by thrombin and consecutive cellular released substances. Indeed, intravenous injections of monomer or implantation of bone cement and prosthesis into the femoral shaft cause acute pulmonary hypertension and trigger a cascade of reduction in gas exchange, right ventricular dilatation, decrease in cardiac output, and drop in peripheral blood pressure. That is, varying concentrations of MMA increase mean pulmonary arterial pressure while decreasing mean arterial pressure in a dose–response manner.^{5,15,25,27,32,57,78,81,90,98-103} Additionally, implantation in vertebral bodies with bone vax versus methacrylate cement increased vascular resistance and caused pulmonary hypertension. ¹⁰⁴

In vitro and vivo studies have shown that MMA accentuates thrombin generation in mixed venous blood and, through co-stimulation, may synergistically contribute to the activation of monocytes, leukocytes, and platelets in adhering and contributing to procoagulant activities and vascular instability.^35,96,97,105 In studies of domestic pigs and other animals, electron micrographs have shown cellular rounding, endothelial gap, and fibrin deposition, all of which reflect the induced thrombotic processes.^82,88 Furthermore, studies in humans found substantially higher levels of central venous thrombin generation and activity following cemented hip implantation when compared to cementless installation. ⁴⁶

Polymerization and Hyperthermia

A high incidence of proximal deep vein thrombosis following hip prosthesis surgery was observed by Stamatakis et al and Planes et al. This correlation was attributed to the hyperthermic state induced by the reactive activation of the coagulation cascade following vascular tissue damage brought on by bone cement curing.^106,107 However, animal experiments showed that the temperature at the interface between the cement and cortical bone scarcely exceeded 40 °C as the metallic implant absorbed and dissipated the thermal energy, thus limiting the activation of coagulation. ¹⁰⁸ These findings were further corroborated in other studies investigating cement curing within the femoral shaft.^35,65 Thus, hyperthermia is unlikely to be the main trigger of the plasma proteolytic system and subsequent intraoperative cardiorespiratory and vascular instability.

Mortality and Current Death Rate

Severe intraoperative hemodynamic instability and hypoxemia have been reported in up to half of cemented cases while uncemented cases tend to mainly exhibit minor hemodynamic instabilities.^4,5,102,103 Nevertheless, although respiratory and hemodynamic impairment are common following cementation of femoral prostheses, sudden fatal outcomes are infrequent.^{5,12,15,16,19,59}

In randomized trials comparing long-term mortality among cemented and uncemented implants, Liu et al reported in a meta-analysis, only sporadic perioperative fatalities (≤48 h) in the cemented groups and no mortalities in the uncemented groups. ¹⁰⁹ However, these trials were all underpowered as it is not entirely feasible to adequately investigate the contribution of cement to mortality due to the dynamics of its physiological action. Nevertheless, 5 national registries on hip fracture patients have presented sufficiently high numbers to study perioperative deaths with analysis demonstrating an approximate 60% increase in early morality (≤ 48 h) among cemented versus uncemented patients (P < .001). ¹¹⁰ Analysis of pooled data found the number needed to harm (NNH) for patients treated with a cemented implant to be 183. ¹¹⁰ In a single, national registry study in which all patients were American Society of Anaesthesiologists (ASA) graded by the operating surgeons, the NNH increased from 116 overall for patients graded ASA 1 to 5 to 33 for patients graded either ASA 4 or 5. ¹¹¹ Registry data included in the meta-analysis revealed that uncemented prostheses had been preferred in weak, elderly patients while the most robust patients received cemented implants. In conjunction with cementation causing significantly higher mortality than uncemented, this divergence in surgical candidates suggests an even larger difference in mortality rates than what has been reported. ¹¹⁰ A subsequent registry-based study corroborated this notion. ¹¹²

Prophylaxis

To mitigate the risk of fatal outcomes following cemented hip joint prosthesis surgery, several preventive surgical and pharmaceutical methods have been employed to modify some of the chemo-cellular processes involved in both the local and systemic coagulation processes. Additionally, prophylactic measures such as high-pressure irrigation of the broached femoral channel, distal plugs, venting catheters, distal cortical drill holes, lavage, bone-vacuum suctioning, and secondary cementation have been studied and found to reduce surgical trauma.^113-118 Despite these prophylactic measures, installation of hip prostheses without methacrylate cement remains the approach of choice in comorbid patients.^110-112

Other prophylactic measures, such as pre-treatment with heparin or a thrombin inhibitor, neutralize most of the coagulation activity and reduce mortality.^42,46,119 These measures further help reduce adverse events through the neutralization of activated, entrapped blood cells in the pulmonary vasculature, thus curbing the pressor response frequently induced by cemented hip joint prosthesis. ⁶⁷ Additionally, the infusion of dextrans improves blood flow and reduces platelet adhesiveness, incidence of pulmonary embolisms, and overall mortality rate in patients. ¹²⁰ Overall, preoperative administered dextrans, heparin derivatives, thrombin, and coagulation factor Xa inhibitors are well-established anticoagulants that prevent catalysis of the coagulation system and reduce mortality.^{42,46,119-121} Clinical trials on antithrombotics in orthopedics have mainly focused on peripheral deep vein thrombosis and neglected the intraoperative systemic hypercoagulation critical to susceptible patients. ¹²²

Summary

Almost 75 years of bone cement research has generated marked advancements in our understanding of the pathophysiology inherent to using cement as an anchor for hip-implanted prostheses.

The main triggers leading to severe intra- and perioperative cardiorespiratory and vascular instability and occasionally death seem to be a moderate neurogenic reflex, marked microembolization, and massive hypercoagulation caused by destruction of the shaft bone marrow and exacerbated by the subsequent impaction of MMA.

Overall, local and systemic hypercoagulation are the main processes in chemo-cellular trafficking that lead to vascular instability and potentially severe sequelae. The most prominent candidates leading to clotting, hemodynamic instability, and organ damage are thrombin, MMA, and substances released from destroyed cells. A multitude of symptoms such as respiratory distress, cardiac arrhythmias, heart failure, angina, myocardial infarction, altered mental status (cognitive symptoms), and paresis have been extensively documented. In a subset of patients, the pathophysiological process continues after surgery, gains momentum, and eventually causes severe adverse effects (vide supra) several days to weeks later. Regardless, cementless prostheses should be preferred in comorbid, elderly patients with thrombin activity controlled via anticoagulation before the critical phase of surgery so as to mitigate the intraoperative coagulation process.

Taken together, although “the bone cement hypercoagulation syndrome” has the potential to invoke severe adverse events and even death through its induced hypercoagulable state, our increasing understanding of its triggers and the utility of prophylactic measures for its prevention has allowed for significant improvements in patient management and overall outcomes.