Damage Control Orthopaedic Surgery: A Strategy for the Orthopaedic Care of the Critically Injured Patient

Early Total Care

In the mid-20th century, early manipulation of long bone fractures was considered unsafe. Because of concerns that fracture manipulation and fixation would increase the incidence of fat emboli, these patients were considered “too sick” for surgery. Traction, prolonged bed rest, and delayed operative interventions were therefore the typical and accepted practice for long bone fractures. In 1967 Kuntscher recommended: “Do not nail immediately, but wait a few days,” with special precautions for patients with multiple fractures or evidence of fat emboli. Others advocated delays of up to 14 days before intervention. Predictably, the consequences were severe. As noted by Border, “traction produces an obtunded patient in the enforced supine position,” which led to deleterious pulmonary effects, poor functional results, and high mortality rates.

In the 1980s, delayed care of femur fractures began to be supplanted by early or immediate fixation. Theorizing that fat embolization was an ongoing process that continued until fracture fixation, Riska et al. of Finland, from the Arbeitsgemeinschaft für Osteosynthesefragen (AO Foundation), first demonstrated improved outcomes with early fixation. In their retrospective series, 22% of patients in the non- or late-stabilization group developed fat emboli syndrome (FES) compared with only 4.5% in the early stabilization group. With the primary goal of improving functional recovery, the AO group also actively promoted early fixation for closed fractures, considering fracture patients “too sick not to be treated surgically.” Such an approach was subsequently supported by multiple outcome studies (10 retrospective, 1 prospective) that focused on the relationship between the timing of fixation of fractures and associated morbidity or mortality. Among the retrospective studies, LaDuca and colleagues extended the AO approach to open fractures using plate osteosynthesis and reported no episodes of FES or cardiopulmonary failure. Describing the experience of the Border group, Seibel et al. noted increased intensive care unit (ICU) stays with delayed stabilization of femur or acetabular fractures. Early fixation reduced the risk of complications, with the caveat, “as long as they are done correctly and in the presence of good oxygen transport and blood clotting.” In contrast, traction significantly increased the cost of care and risk of multiple system organ failure. Goris and colleagues reported their experience in the Netherlands, where reductions in mortality and acute respiratory distress syndrome (ARDS) were realized with early plate osteosynthesis. Johnson and colleagues similarly reported a fivefold increase in ARDS and mortality rate if fixation was delayed beyond 24 hours. This was most pronounced for the severely injured (Injury Severity Score [ISS] >40).

The first randomized, prospective study to support early fixation of long bone fractures was conducted by Bone and colleagues at Parkland Memorial Hospital, Dallas, Texas. In this landmark study, 83 patients with femoral shaft fractures and ISS of 18 or greater were divided into early-fixation (<24 hours) and late-fixation (>48 hours) groups. In the 46 patients in the early-fixation group, a total of 16 pulmonary complications were observed (ARDS, pulmonary embolism [PE], fat emboli syndrome, or abnormal blood gases), including only one case of ARDS. In the 37 patients in the late-fixation group, 50 pulmonary complications were observed, with six cases of ARDS. These results were considered a confirmation of the findings of the previous retrospective studies, and resulted in “the overwhelming recommendation … that early stabilization of long bone fractures should be performed in multiply injured patients.” ETC became the standard of care, and practice patterns changed. In Border's group, the average duration for traction before fixation of femoral shaft fractures decreased from 9 to 2 days. Today, the Bone study is viewed with only slightly less enthusiasm. The only statistically significant difference presented was an increase in total hospital costs for the delayed-fixation group. The more clinically relevant findings exhibited by the late group—increased pulmonary complications and longer hospital and ICU stays—were merely trends that failed to reach statistical significance. In addition, the randomization process was somewhat flawed, resulting in associated pulmonary injuries in 10 of 37 patients in the delayed-fixation group compared with only 1 of 46 in the early group.

Currently, it is near universally accepted that ETC, defined as the definitive fixation of fractures within approximately the first 24 hours after injury, is appropriate for the vast majority of patients. When contrasted and compared with treatment of femur fractures with prolonged traction and delayed fixation, ETC avoids the deleterious effects of prolonged recumbence and is associated with improved survival and decreased complication rates.

Damage-Control Orthopaedics

The term damage control has its origins in the maritime industry. It is a shipboard doctrine applied in an effort to limit damage (from fire, flooding, or other hazard) to a confined area of an afflicted ship, thereby maximizing its overall survivability. This is achieved by anticipating and rapidly containing the cascade of after-effects of a potentially catastrophic event. Simply stated, it is an effort to “keep the ship afloat.”

In the early 1980s, general surgeons began the practice of early termination of emergent laparotomies for penetrating trauma upon completion of lifesaving interventions. Appreciating the similarities that this approach shared with shipboard damage-control efforts practiced by the US Navy, Rotondo et al. first used the term damage control to describe this laparotomy management tactic in 1993. With similar goals in mind for trauma patients, surgical damage control consists of three phases, as described by Feliciano and colleagues:

• 1.

  Initial operative intervention for control of life-threatening bleeding and decontamination

• 2.

  Intensive care transfer for correction of the deadly triad of hypothermia, acidosis, and coagulopathy

• 3.

  Return to the operating room for definitive repair of the intra-abdominal injuries.

Better-than-expected survival rates have been realized with this strategy, which remains popular and effective today. Eventually, damage-control strategies were developed for the initial treatment of extraabdominal trauma to include the musculoskeletal system. The damage-control concept as applied to long bone fractures was introduced under the moniker of “damage-control orthopaedics,” as coined by Scalea and colleagues.

In the 1990s, data from the trauma service in Hannover, Germany, and other studies suggested that ETC could have negative effects in certain subsets of the MIP population, namely, those with severe pulmonary and head injuries. For these individuals, the initial treatment of fractures with provisional external fixation, followed by delayed definitive fixation, has potential advantages. Similar to ETC, such an approach avoids the deleterious effects of prolonged recumbence, but in contrast, it spares the patient from the additional physiologic burden of major surgery soon after injury, particularly the embolization of fat and marrow contents associated with intramedullary instrumentation that has been associated with secondary injury to the pulmonary capillary endothelium.

In their 1993 study, the Hannover group raised concerns that early (<24 hours) femur fracture fixation with reamed intramedullary nailing (IMN) was associated with deleterious pulmonary effects and higher mortality in patients with concomitant chest injuries. They retrospectively reviewed 106 patients with femur fractures and an ISS ≥ 18, splitting them into four groups based on the presence or absence of a severe chest injury and early (<24 hours) or late (>24 hours) IMN fixation. Consistent with other studies, ETC was noted to be advantageous in the non–chest-injured group. In the chest-injured group, however, they found a higher (although not statistically significant) incidence of ARDS in the early-IMN group (33%) compared with the late-IMN group (7.7%). They concluded that “in the presence of pulmonary injury, primary intramedullary femoral nailing causes additional pulmonary injury and may trigger ARDS.” Critics of the study point to the small group size; lack of statistical significance; a higher incidence of ARDS than reported in the North American literature; and the fact that among the chest-injured patients, 29% of the early-IMN group had bilateral chest injuries, versus only 7.7% in the late-IMN group. The magnitude of the chest injury, a factor poorly considered by the Abbreviated Injury Scale (AIS) scoring system, has been associated elsewhere with significant clinical importance. Regardless, the findings were concerning and led to efforts in the 1990s to avoid early reamed IMN in this subset of patients.

ARDS and multiple organ failure (MOF) are the dreaded endpoints of the systemic inflammatory response syndrome (SIRS), an exaggerated inflammatory response that ultimately damages organ systems that may have been uninvolved in the initial trauma. There are two described inflammatory models for SIRS. In the “one-hit” model, a massive initial injury and shock incite SIRS, resulting in early end-organ injury. In the “two-hit” model, an initial injury or “first hit” incites an appropriately heightened state of SIRS that, if followed by a “second hit,” can be amplified and result in late MOF. Second hits may be from severe hemorrhage, incomplete resuscitation, infection, or major surgery. Of concern for this MIP subset was the potential for the timing and/or method of fixation to act as a second hit.

DCO is generally conducted by immobilizing a long bone or pelvic fracture with a temporizing external fixator to achieve the advantages associated with ETC (fracture stability, decreased pain, ease of nursing care, improved patient positioning in the ICU, decreased fat emboli ) while minimizing the potential adverse effects of major surgery (blood loss, hypothermia, and inflammatory system stimulation), which could serve as a second hit. After adequate resolution of physiologic stability, and once the patient is no longer hyper-susceptible to the second hit of surgery, definitive fixation is conducted ( Fig. 12.1 ). Avoiding the consequences of an exaggerated second hit and the development of the lethal triad is the goal of DCO. This far outweighs the major disadvantages of external fixation, which include the need for additional surgical procedures, the increased cost, and the potential increased infection risk with prolonged application.

Identification of the subset of patients who would benefit from DCO is a continuing process, generating much controversy because indiscriminate application of DCO strategies could actually be harmful in addition to incurring unnecessary expense. The generation of all-inclusive algorithms and strictly defined indications and treatment recommendations for the orthopaedic management of MIPs has proven elusive. In 2005 Rixen and colleagues reviewed 63 controlled DCO trials and failed to find a “generalized management strategy.” Even proponents admit that considerable clinical judgment and experience remain prerequisites to the appropriate application of DCO. Despite considerable experience, published DCO implementation rates varied dramatically among highly regarded institutions—12% at the University of Maryland's R. Adams Cowley Shock Trauma Center (2002–2005) versus 57% at Denver Health Medical Center (1993–2006) for similarly described patient populations. Improved understanding of the inflammatory process and the importance of resuscitation have been critical in better defining the appropriate DCO population.

Although DCO protocols have been implemented at most trauma centers in the United States and Europe, the specific clinical indications remain unclear. Unfortunately, much of the DCO literature describes studies of limited scientific and statistical power because of the complex nature of this patient population. Difficulties in the study of these patients include their dynamic and often complicated clinical course, quantification of the impact of associated injuries, and an evolving understanding of the underlying pathophysiology of the inflammatory process. Confounding factors include multiple significant trauma care improvements in resuscitation and ventilation, ICU clinical advances that occurred concurrently with the implementation of DCO protocols, and potential underlying differences in the populations and geographic variation in resuscitative protocols.

In addition, the definition of DCO has gradually expanded in some studies to include non-MIP patients, further confusing the literature. At first, the term DCO described the use of femoral external fixation in an MIP as a temporizing measure—or bridge—to allow appropriate resuscitation before the definitive procedure. Over time, a change in this definition has resulted in the use of the term DCO to describe external fixation of relatively isolated extremity fractures to provide temporary stabilization, typically to allow for resolution of soft tissue swelling or local wound healing, or to protect a vascular repair, before definitive fracture fixation. Clearly, these patients differ dramatically from the MIP in need of systemic resuscitation to allow for safe definitive fracture treatment. It is therefore useful to adopt the term limb damage control (LDC), as utilized by Roberts et al., to differentiate between these diverse patient populations and treatment strategies. The local advantages of LDC are well documented and discussed elsewhere in this text (i.e., periarticular tibial plateau or pilon fractures). The systemic advantages of DCO are less clear.

Early Appropriate Care

The introduction of DCO was met with some degree of controversy, particularly in North America. This has continued to the present day as a result of the lack of demonstrated benefits after implementation of DCO protocols, particularly with respect to ARDS. Pape has warned against “the transfer of elements from one trauma system to the other without careful consideration,” noting deleterious results in Central Europe when ETC guidelines from the US literature were widely applied there. Although not deleterious, the implementation of DCO guidelines in the United States has not fully realized the benefits that were anticipated from the European literature. Potential reasons for the regional disparities are differences in trauma systems, resuscitation practices, and population characteristics or that DCO has been overdone. The retrospective study of Pate et al. in 2002 comparing outcomes during the ETC, intermediate, and DCO protocol eras practiced in Hannover, Germany, noted an ARDS rate of 25% during the DCO era, an improvement from the 58% rate during the ETC era. In contrast, in 2009, O'Toole et al. published the Baltimore Shock Trauma experience with a similar population, in which DCO was used less frequently, yet they found an ARDS rate of only 1.5%. In 2012, Nahm and Vallier published the Cleveland MetroHealth Medical Center experience, finding a similar low ARDS rate of 1.7%. Further concerns regarding the applicability of the Pape et al. 2002 study were raised by the 2007 EPOFF study from Europe, in which patients were randomized to early IMN (ETC) versus external fixation (DCO). In “borderline patients,” early IMN demonstrated only an increased incidence of acute lung injury (ALI), but no increase in pneumonia, ARDS, ventilator days, or SIRS. These concerns prompted the introduction in 2011 of an Early Appropriate Care (EAC) protocol by the Cleveland group in which femoral fractures are preferentially treated in the first 24 hours (similar to ETC), whereas other extremity fractures are splinted and fixed at a later date (in contrast to ETC), thus providing somewhat of a compromise between the ETC and DCO protocols. Using readily available acid–base system laboratory values as an indicator of resuscitation status ( Box 12.1 ) and a team-based surgical clearance approach as is typical in North America (led by the general surgery/trauma service), the implementation of the EAC protocol resulted in fewer complications and shorter hospital stays. Per the authors, if any one of the initial acid–base parameters improved to the listed levels during the first 36 hours, and clearance was provided by the trauma service, definitive management of the femur, pelvis, acetabulum, or thoracolumbar spine could be safely conducted. The study is underpowered and has been criticized for its reliance on the acid–base system alone for clearance, which could provide an inaccurate indication of the true resuscitation status in diabetics and those with chronic renal failure or severe liver trauma. Parameters assessing coagulopathy, clotting factors, and circulatory and pulmonary functions are also not specifically addressed in the EAC protocol, although these are likely embedded in the trauma surgery clearance process. Clearly, easily accessed data, which can be repeated at intervals, are attractive for use in distinguishing those patients most appropriate for DCO versus EAC/ETC. Recognizing this, Pape and Pfeifer in 2015 coined the term safe definitive orthopaedic surgery (SDS), which describes clinical criteria that can be evaluated at intervals for selection of DCO versus EAC/ETC in borderline patients ( Fig. 12.2 ).

Resuscitation