Orthopedics

Review Article 

Thigh Pain After Total Hip Replacement: A Pathophysiological Review and a Comprehensive Classification

Luca M.C. Pierannunzii, MD

  • Orthopedics. 2008;31(7)
  • Posted July 1, 2008

Abstract

Total hip replacement (THR) is a successful procedure that provides excellent pain relief and considerable improvement of function. The postoperative satisfaction rate is estimated to be as high as 90%.1

With regard to the dissatisfied 10%, pain is the main factor that makes surgery fall short of the patient’s expectations. Notwithstanding the main source of pain is loosening, painful well-fixed THRs are not exceptional: infections, radiculopathies, trochanteric or iliopsoas bursitis are some of the most common diagnoses the physician has to exclude.2

When all, particularly the incipient infection, have been ruled out, thigh pain should be considered.

Thigh pain is a symptom, not a disease. In the typical presentation it is localized approximately mid-thigh, in the anterolateral aspect of the limb, grossly at the level of the tip of the prosthetic stem. If the patient reports inguinal or posterior localization, other diagnoses should be considered.

Because it depends on the mechanical load of the implant, it is absent at rest, reaches its peak in the start-up phase, and diminishes over the movement.

The frequency shows high variability, ranging from <1% to 40%.3-5

The clinical management remains a challenge, since no medical treatments allow long-lasting relief and no surgical treatments are sustained by adequate evidence.

The end-of-stem pain, as this symptom is also named, is a multifactorial phenomenon, in which often a single cause cannot be clearly identified.6

This article reviews the main theories about thigh pain pathogenesis. A pathophysiological classification, consistent with the described models, is presented with the goal of providing surgeons with a practical tool for clinical decision-making.

Although many hypotheses have been proposed about end-of-stem pain generation, most can be easily traced back to one of the following mechanisms: tip micromotions and tip overload.

In the former, the end of the stem moves inside the femoral canal over the gait cycle because of the cyclic load of the joint, so determining an oscillating stimulus of the endosteum, that is densely innervated.

In the latter, the stem does not transmit the applied load to the femur along its full length, but concentrates it around the tip, through a phenomenon known as stress transfer. This localized overload produces excessive bone strain and endoperiosteal irritation, together with some important densitometric and morphological changes of the proximal femur.

Cemented THRs are rarely associated with thigh pain because bone cement both prevents from apical oscillations and spreads the load over the full length of the stem.5,7 Thus, when a patient who had a cemented THR reports femoral pain, the most common diagnosis is frank loosening.8

Two different conditions may allow micromotions of the tip: stem loosening and fibrous fixation.

Stem Loosening

The main clinical symptom of a loose femoral component is thigh pain.9-11

Although radiological features of frankly loose stems (subsidence, pedestal formation, cortical hypertrophy, and increasing radiolucencies) allow easy diagnosis of the condition, the incipient loosening is barely demonstrated on plain radiographs, since the morphological changes are subtle and not easily recognized recognized.

99Tc methilene diphosphonate (99Tc MDP) bone scan is an effective tool in early identification of the mechanical failure, although it does not permit any clear distinction between septic and aseptic loosening.12

Sometimes the stem appears to be loose early; better defined as “primarily unstable,” ie, the component has never achieved a biologic fixation in the bone. These cases are important diagnostic challenges, as they might conceal an occult biofilm-associated infection. Often routine diagnostic approach fails in detecting those infections, in which ESR, CRP and scintigraphic assays may turn out to be false negative.13

Fibrous Fixation

Grossly fixed stems may permit oscillations of the tip if stability is not…

Total hip replacement (THR) is a successful procedure that provides excellent pain relief and considerable improvement of function. The postoperative satisfaction rate is estimated to be as high as 90%.1

With regard to the dissatisfied 10%, pain is the main factor that makes surgery fall short of the patient’s expectations. Notwithstanding the main source of pain is loosening, painful well-fixed THRs are not exceptional: infections, radiculopathies, trochanteric or iliopsoas bursitis are some of the most common diagnoses the physician has to exclude.2

When all, particularly the incipient infection, have been ruled out, thigh pain should be considered.

Thigh pain is a symptom, not a disease. In the typical presentation it is localized approximately mid-thigh, in the anterolateral aspect of the limb, grossly at the level of the tip of the prosthetic stem. If the patient reports inguinal or posterior localization, other diagnoses should be considered.

Because it depends on the mechanical load of the implant, it is absent at rest, reaches its peak in the start-up phase, and diminishes over the movement.

The frequency shows high variability, ranging from <1% to 40%.3-5

The clinical management remains a challenge, since no medical treatments allow long-lasting relief and no surgical treatments are sustained by adequate evidence.

The end-of-stem pain, as this symptom is also named, is a multifactorial phenomenon, in which often a single cause cannot be clearly identified.6

This article reviews the main theories about thigh pain pathogenesis. A pathophysiological classification, consistent with the described models, is presented with the goal of providing surgeons with a practical tool for clinical decision-making.

Etiopathogenetic Models

Although many hypotheses have been proposed about end-of-stem pain generation, most can be easily traced back to one of the following mechanisms: tip micromotions and tip overload.

In the former, the end of the stem moves inside the femoral canal over the gait cycle because of the cyclic load of the joint, so determining an oscillating stimulus of the endosteum, that is densely innervated.

In the latter, the stem does not transmit the applied load to the femur along its full length, but concentrates it around the tip, through a phenomenon known as stress transfer. This localized overload produces excessive bone strain and endoperiosteal irritation, together with some important densitometric and morphological changes of the proximal femur.

Cemented THRs are rarely associated with thigh pain because bone cement both prevents from apical oscillations and spreads the load over the full length of the stem.5,7 Thus, when a patient who had a cemented THR reports femoral pain, the most common diagnosis is frank loosening.8

Micromotions

Two different conditions may allow micromotions of the tip: stem loosening and fibrous fixation.

Stem Loosening

The main clinical symptom of a loose femoral component is thigh pain.9-11

Although radiological features of frankly loose stems (subsidence, pedestal formation, cortical hypertrophy, and increasing radiolucencies) allow easy diagnosis of the condition, the incipient loosening is barely demonstrated on plain radiographs, since the morphological changes are subtle and not easily recognized recognized.

99Tc methilene diphosphonate (99Tc MDP) bone scan is an effective tool in early identification of the mechanical failure, although it does not permit any clear distinction between septic and aseptic loosening.12

Sometimes the stem appears to be loose early; better defined as “primarily unstable,” ie, the component has never achieved a biologic fixation in the bone. These cases are important diagnostic challenges, as they might conceal an occult biofilm-associated infection. Often routine diagnostic approach fails in detecting those infections, in which ESR, CRP and scintigraphic assays may turn out to be false negative.13

Fibrous Fixation

Grossly fixed stems may permit oscillations of the tip if stability is not absolute. This situation has been related to the quality of tissue ingrowth on the porous-coated surface.

If the primary stability of a porous-coated prosthetic component is high, within a few months it is likely to develop a significant bone ingrowth. Bone ingrowth is radiographically shown by no subsidence of the stem, direct bone-prosthesis contact, and several spot-welds.14,15

When those features were visible, Engh et al14 retrieved thigh pain as rarely as in 10% of patients who had received a cylindrical Co-Cr stem (AML, DePuy, Warsaw, Indiana).

When wider micromotions are allowed immediately after implantation, the tissue ingrowth is mainly fibrous. This kind of fixation is radiographically visible, as a thin nonevolutive radiolucent line (<1 mm), externally enveloped by a radiopaque zone, surrounds the prosthesis. Minimal subsidence is often observed if compared to the first postoperative radiograph (Figure 1). Subsidence, if any, appears stable at subsequent controls, as fibrous ingrowth stops further movements at bone-prosthesis interface. The rate of end-of-stem pain reported with the above prosthesis grows up to 28% among the fibrous-stabilized implants. According to the same authors, the different incidence of thigh pain should be related to the different secondary stability respectively ensured by the two kinds of biologic fixation.

The extent of porous coating is an important factor governing the secondary fixation of cementless arthroplasties: the wider the coated part of the stem, the higher the biologic stability. According to Engh and Bobyn’s case series,16 fully coated femoral components show important bone ingrowth and low thigh pain incidences, but determine significant stress shielding. Proximally coated stems, however, would be more likely to determine midthigh discomfort, but would be rarely associated with proximal bone resorption. From this point of view, end-of-stem pain and stress shielding seem to be specular problems, respectively due to defect and excess of distal fixation. This view disregards the overload-related pain often associated with stress shielding phenomenon.

In addition to the extent of porous coating, canal fill is another factor determining implant stability, and thus thigh pain risk.

Engh et al14 reported that bone ingrowth developed in 93% of the hips when the isthmus was tightly filled, but only in 69% when it was loosely filled (P<.005). In the first group, thigh pain rate was as low as 9%, while in the second it was 23% (P<.05).

Similar results are reported by Whiteside,17 who again demonstrated that the greater is the distal fill, the lower is the incidence of thigh pain. Using a cylindrical collared femoral component in 105 THR, he retrieved a thigh pain rate of 3% among distally tight stems and the much higher rate of 53% among distally loose stems.

Those data seem to confirm that the implantation of fully coated conventional stems completely filling the medullary canal reduces the chance of end-of-stem pain.

Figure 1A: Thigh pain determined by fibrous ingrowth around a fully coated tapered stem with poor distal fit Figure 1B: Eighteen months later thin radiolucent lines, externally lined by more radiopaque bone, are visible along the whole lateral aspect of the stem and the proximal part of the medial aspectFigure 1C: Eighteen months later, the 99Tc MDP bone scan shows mild focal uptake around the tip
Figure 1: Thigh pain determined by fibrous ingrowth around a fully coated tapered stem with poor distal fit. Early postoperative radiograph (A). Eighteen months later thin radiolucent lines, externally lined by more radiopaque bone, are visible along the whole lateral aspect of the stem and the proximal part of the medial aspect; slight subsidence may be observed over the shoulder of the prosthesis (B). Eighteen months later, the 99Tc MDP bone scan shows mild focal uptake around the tip (C).

As for neck-preserving stems, data are few. Pipino et al18 reported thigh pain with Biodynamic Hip (Howmedica, Rutherford, New Jersey), a macroporous-surfaced Co-Cr comma-shaped anatomical stem. In this case series, 6 of 44 patients reported femoral pain. Interestingly 1 of 3 oversized stems (33.3%), 5 of 19 undersized stems (26.3%), but none of 22 properly sized stems were painful. Stems were considered undersized if the distance between the distal third and the cortex was wider than 2 mm, oversized if there was direct contact.

Pipino et al18 reported a self-resolving trend of the symptoms within 1 year among undersized implants, which is consistent with the theory of initial defect of stability, subsequently followed by bone ingrowth. The radiological finding of transient thin radiolucent lines seems to confirm this hypothesis, according to which high intrinsic stability of neck-preserving anatomical designs might overcome undersizing problems, allowing delayed bone ingrowth even in deficient canal fill.19

Overload

Excessive stress transfer to the apex of the stem may determine pain as well as its mobility. Many factors are involved in stress distribution: the difference of rigidity between implant (as combination of design, materials and size) and host bone, the extent of porous coating, and possibly the alignment.

Rigidity Mismatch

According to Koch’s model of hip biomechanics, load application to the proximal femur, given its offset structure, determines flexion, which results in compression forces along the medial aspect and distraction forces along the lateral aspect.20

When a solid sustains a bending deformation, the relationship between the applied bending moment Γ and the resulting deformation (ie, the acquired radius of curvature R) is:

Γ=ρf/R

where ρf is the flexural rigidity, a physic measure of the bending stiffness. It defines the moment required to bend a rod or a beam to a unit radius of curvature. The flexural rigidity is calculated as the product E Ia, where E is the modulus of elasticity and depends only on the material, and Ia is the area moment of inertia and depends only on the cross-section.

Each material is characterized by a specific modulus of elasticity. The higher the modulus, the stiffer the material. Moduli of diaphyseal bone, titanium alloy, and cobalt-chromium are approximately 12 to 16 GPa, 110 GPa and 220 GPa respectively.

Figure 2A: Osteoporosis-related thigh pain in an 80-year-old patient. Figure 2B: Two years later, cortical hypertrophy around the tip and huge trochanteric resorption attest focal stress concentrationFigure 2C: Two years later, the high 99Tc MDP uptake around the tip confirms the radiological diagnosis of tip overload
Figure 2: Osteoporosis-related thigh pain in an 80-year-old patient. The tight distal fill excludes any micromotion, but the great rigidity mismatch between the large component and the osteoporotic shaft determines severe tip overload. Early post-operative radiograph (A). Two years later, cortical hypertrophy around the tip and huge trochanteric resorption attest focal stress concentration (B). Two years later, the high 99Tc MDP uptake around the tip confirms the radiological diagnosis of tip overload (C).

However, the area moment of inertia represents the “geometrical resistance” to bending and is influenced by shape and area of the cross-section.

In the easiest case of solid tube with circular section (ie, cylindrical stem), it is equal to:

Ia=¼ π r4

where r is the radius. As the radius is considered to the power of 4, a small size increase may augment the area moment.

If the section is rectangular (ie, tapered squared-section stem), the relationship becomes:

Ia=a3 b /12

where a is the dimension that lies in the plane of deflection (height) and b is the dimension perpendicular to the plane of deflection (width). Noticeably the height, being raised to the power of 3, is more important than the width. This means that Zweymuller’s design leads to high area moment of inertia notwithstanding the blade-like shape, since the minor dimension is orthogonal to the cervical plane (plane of deflection) and then is not raised, while the major dimension is cubed.

In the more complex case of hollow tube with circular section (ie, femoral diaphysis), the relation considers both inner (r) and outer (R) radii, becoming:

Ia=¼ π (R4-r4)

Since the femoral size is bigger than any stem size, regardless of the design, the area moment of inertia of the former is generally bigger than that of the latter. Skinner et al,21 reviewing 101 THRs, found that the femur area moment was 22.4 times bigger on average than the area moment of the properly sized Harris-Galante stem (Zimmer, Warsaw, Indiana), 37.6 times bigger than the AML stem, 19.2 times bigger than the PCA stem (Howmedica).

The lower area moment of inertia of prosthetic components compensates their higher modulus of elasticity. For this reason the flexural rigidity of the femur is higher than the flexural rigidity of the stem, even if cobalt-chromium alloy is chosen. This conclusion was confirmed by the flexural rigidity bone-to-implant ratio, which is higher than 1 in any case, on average 3.36 for Harris-Galante system, 2.65 for AML, 1.36 for PCA. Skinner et al21 could not demonstrate a significant correlation between the bone-to-implant ratio of flexural rigidity and pain, although their data might suggest a trend towards lower pain with more flexible prosthesis.

Burkart et al22 reported a lower thigh pain rate with titanium alloy than with cobalt-chromium, but their results hardly demonstrate the role of bending stiffness, as the 2 studied stems were different. Lavernia et al23 found no difference between titanium-made and cobalt-chromium-made versions of the same prosthesis. These studies should reduce the emphasis placed on flexural rigidity in generation of thigh pain.6,24

The importance of flexural rigidity in stress distribution along the proximal femur was introduced in 1987 by Engh and Bobyn.16 At that time the human hip biomechanics was represented by Koch’s model, published at the beginning of the 20th century.20 That model, applicable if the joint was loaded without any input from the soft tissues, does not describe the physiological conditions in which muscles play a critical role balancing the bending moment determined by body weight.25,26

In light of the tension band effect of the iliotibial tract, lateral cortex is expected to go through compression forces as well as the medial one during unilateral support phase of gait. According to this new model, the main bending moment acting on proximal femur is neutralized by muscular power. Although ambulation, stair climbing, standing up, and other common activities may determine other unbalanced bending moments, their role should be put in perspective.

Other deformations could play a major role, such as axial compression and torsion.

In the case of torsion, the relationship between the applied torque T and resulting torsional deformation α/l (where α is the angle of torsion and l the length) is:

T=ρp α/l

where ρp is the torsional rigidity, which is the moment required to twist a rod of unit length over a unit angle (1 rad=57.3°). Also the torsional rigidity can be calculated as product of a modulus–shear modulus G–and a moment of inertia–polar moment of inertia Ip. Similarly G depends only on the material and Ip on the shape and area of the cross-section.

In pure compression, the relation between the compressive force F and the resulting relative shortening Δl/l (where l is the initial length and Δl its variation) is:

F=ρa Δl/l

where ρa is the axial rigidity. Differently from flexural and torsional rigidity, the axial one does not depend on the shape of the cross-section, but only on its area (A) and its modulus of elasticity (E). It is directly obtained by the product E A.

Although the correlation between flexural rigidity mismatch and thigh pain appears to be weak, to our knowledge no studies have addressed these types of rigidity.

However, regardless of the kind of deformation, if the proximal femur receives a prosthetic stem and a firm bond, such as bone ingrowth, develops between them, its overall rigidity is supposed to increase up to the sum of native femoral rigidity and acquired implant rigidity.16 Under the tip of the stem, the femur is no longer stiffened and is supposed to return to its original elastic properties.

Since higher rigidity leads to lower deformation, a concentration of bone strain is expected at the end of the femoral component. As periosteum and endosteum are densely innervated, the hyperstrain localized at the tip might determine nociceptors stimulation.27 This should be the basis of thigh pain secondary to stiff stem.

Furthermore, lower deformation leads to lower stress. A tightly bound stem determines stress protection along its length and stress concentration at its tip.28

The higher its rigidity (or the lower the bony shaft rigidity), the greater this effect, globally known as stress shielding.

According to Wolff’s law, remodeling is directed by stress distribution, resulting in resorption where no stresses are transmitted and cortical hypertrophy where excessive stress is sustained.29 This should be the cause of proximal bone resorption secondary to stiff stem.

The hypothesis of rigidity mismatch is consistent with several risk factors for thigh pain already brought out by clinical observation:

  • high modulus of elasticity of the stem (ie, chromium-cobalt alloy),21,22
  • large size,30
  • solid design (versus grooved or slotted ones),31 and
  • low modulus of elasticity and section area of the femur (ie, osteoporosis).14

Osteoporosis both reduces femoral rigidity and allows the implantation of larger stems, as medullary canal is widened (Figure 2). The result is a noticeable rigidity mismatch, that appears to justify Engh’s findings: pain rate approximately 11% among patients with good preoperative bone quality and 26% among those with poor bone quality.14

Extensive Porous Coating

The extent of porous coating influences the extent of fixation and the amount of stress shielding. The rough calculations discussed above refer to the easy model of a fully osteointegrated stem. In pure proximal fixation, the precise analysis of the elastic behavior of femoral-stem complex goes far beyond the purposes of the present article. However, it is clear that the more proximal the fixation, the higher the femoral-stem complex deformation, the lower the stress shielding.

Radiological findings of stress shielding are more often retrieved with fully coated stems than with proximally coated stems. According to Engh and Bobyn,16 the prevalence of severe bone resorption with one-third coated stems was approximately half the prevalence observed with two-third and fully coated implants (P<.05).

Theoretically fully coated stems are protected from micromotions-related pain, but prone to overload-related pain. It is unclear which phenomenon is more important, as data available from literature seem to show similar pain rates.

On extensively coated stems, McAuley et al32 reported a prevalence of 12% of 381 cylindrical cobalt-chromium femoral components reviewed at 9-year mean follow-up, but only one fourth of those conditions was considered disabling. Naumann et al,33 reviewing 413 titanium-made squared-section Zweymuller stems (Sulzer, Zurich, Switzerland) with short-term follow-up, found 23.9% slightly painful and 5.6% severely painful. Schramm et al,34 following-up for 10 years 87 tapered titanium-made CLS Spotorno stems (Zimmer), retrieved a thigh pain rate of 17%.

Despite the different shape (respectively cylindrical, squared-section blade-like and squared-section wedge-like) and philosophy, that could be summarized as fit-and-fill, fit-without-fill and press-fit, the pain rates are substantially alike.

On the proximally coated stems, a retrospective study of 111 Porous Coated Anatomic (PCA) stems (Howmedica) by Campbell et al,4 showed an end-of-stem pain rate increasing over time, as high as 13% at 1 year and 22% at 2 years. The same stem had even worse results at 6-year follow-up according to Kim and Kim,35 who reported a pain rate as high as 25%.

However, the differences of design and material cannot be underestimated, as the PCA stem is cylindrical, anatomically-shaped and Co-Cr-made. Bourkart et al,22 analyzing the results of 105 Mallory-Head stems (Biomet, Warsaw, Indiana), tapered and titanium-made, found thigh pain only in 7% of patients at 1 year and 3% at 2 years. Lavernia et al23 reported slightly higher pain rates with another tapered design, the Trilock stem (DePuy), regardless of the material: 9.5% after 1 year and 8.7% after 2 years.

Although these studies about proximally coated stems appear to show a lower risk of thigh pain when tapered design is chosen rather than anatomic, it should be noted that the PCA represents a poor combination of many features, such as high flexural rigidity with proximal coating and anatomic shape. Other anatomic designs obtained better results, at least equal to the tapered designs. Rogers et al36 recorded persistent thigh pain only in 3% of 100 Anatomique Benoist Giraud (ABG) total hip arthroplasties (Howmedica) followed-up for 6 years on average.

According to the literature, extensive porous coating does not appear to significantly influence the risk of thigh pain, since it prevents from micromotions, but is likely to enhance distal stress transmission.

Varus Malalignment

The importance of varus malalignment in stress transmission and thigh pain is commonly accepted but poorly demonstrated. Even if Gill et al37 described insufficiency fractures around the tip of stems implanted in varus in osteopenic femora, the evidence of the relationship between varus placement of uncemented stems and those unfavorable consequences is far from being demonstrated.

The catastrophic results of stems cemented in varus are more strictly associated with the insufficient lateral cement mantle than with other mechanical factors.38,39

With regard to cementless THR, Panisello et al40 analyzed the different clinical, radiological and densitometric outcomes of 69 hips according to the orientation of the stem. All the stems were proximally-coated, titanium-made and anatomic (ABG II; Stryker Howmedica, Staines, England).

Although Panisello et al40 retrieved more important atrophy in zone 7 (greater trochanter area) and hypertrophy in zone 6 (lateral cortex at the metadiaphyseal transition) in the varus group than in neutral group, suggesting more distal stress transfer to the lateral cortex in case of varus alignment, these results could hardly represent evidence, as the numerical discrepancy between the 2 samples was as large as 6 to 54 (the remaining 9 stems were valgi). For the same reason, the reported high incidence of thigh pain (2/6) in the varus group should not be overestimated.

However, Rogers et al36 using the precursor of the above stem (ABG), did not show any significant difference in clinical outcome between neutrally aligned and malaligned stems. Khalily and Lester41 found no outcome deterioration between neutrally aligned and varus-aligned extensively coated titanium-made Zweymuller’s stems (Alloclassic; Sulzer). Schneider et al42 had similar conclusions after reviewing 3732 CLS Spotorno stems, another titanium-made extensively coated tapered stem.

The relationship between varus placement and thigh pain is further weakened by the experience of Morrey et al,43 who followed-up 159 Mayo Conservative Hips (Zimmer) for an average 6.5 years. This particular short wedge design was developed to be inserted varus into the proximal femur, until the bent apex reaches the posterolateral cortex. None of the patients reported significant end-of-stem pain.

All the above reports are consistent with the tension band model of hip joint loading proposed by Fetto et al,25,26 in which the bending moment is neutralized by the abductor muscles and especially by the iliotibial tract.

According to this model, moderate varus malalignment would not determine any greater flexion of the femur around the tip and consequently any clinical impairment.

Pathophysiological Classification

Since the two main pathomechanisms leading to thigh pain have been shown to differ from each other, the attempt to classify the disorder according to its pathophysiology seems to be justified.

From an epidemiological point of view, such a distinction could allow more effective reporting of the problem, helping to identify which feature of a prosthetic design, step of a surgical procedure, or criterion of patients selection should be changed to reduce the risk of postoperative pain. Whether the problem arises from micromotions or overload is another matter.

From a clinical point of view, the same distinction would equip the surgeon with a logical tool to determine adequate treatment.

Pursuing the goal of developing a classification both pathophysiology-based and clinically oriented, thigh pain has been divided into dynamic and static forms. The diagnostic algorithm is displayed in Figure 3.

Figure 3: Diagnostic algorithm. Abbreviations: Neg, negative; pos, positive
Figure 3: Diagnostic algorithm. Abbreviations: Neg, negative; pos, positive.

Dynamic thigh pain is related to micromotions of the tip, caused by gross loosening (macroinstability) or weak proximal fixation such as fibrous ingrowth (microinstability).

On the other hand, static thigh pain is related to stress shielding, particularly to the overload of the tip. While distal fixation and rigidity mismatch appear to be certain factors, varus malalignment has not been proved as well.

The clinical importance of a classification system depends on its possibility to guide the therapeutic choice. No precise guidelines have been determined about the management of painful not loosened THR.

Brown et al6 suggest a “watch and wait” approach, as some authors documented a decreasing trend of symptoms over 2 years.22-44 During the watchful waiting phase, the recommendations are summarized in four As: activity modification, ambulatory assistive devices, appropriate anti-inflammatory medications, and adequate time for nonsurgical treatment.

Because thigh pain is essentially load-related, another A should be added to the previous list: appropriate weight loss. Skinner et al21 reported that the flexural rigidity ratio did not correlate significantly with pain score if considered alone, but when it was normalized by dividing by patients’ weight, it showed a significant correlation.

After the “watch and wait” period, the likelihood of spontaneous regression is minimal. In this case, the pain level should guide any further decision. Only disabling pain should be considered for surgical treatment, as it requires invasive procedures that cannot be justified only for mild discomfort.

Two main surgical options have been described: revision of the stem and strut grafting of the femur.6,24

Although the two procedures are different, no clear differential indications have been proposed nor have clinical trials been reported. In the absence of any evidence, the above classification might lead to a reasonable therapeutic choice.

Dynamic pain, arising from mechanical instability of the femoral component needs to be addressed through stem substitution to achieve a stable fixation. Cementing should be considered if good primary stability cannot be guaranteed otherwise, as the protection from thigh pain assured by cement fixation is well known.45

Unfortunately, the long-term survival rate of cemented revisions after failure of uncemented implants is fair, and possibly worse than cementless techniques.46 However, uncemented revisions are frequently complicated by thigh pain, usually static, because of the implantation of rigid large-sized fully coated stems in poor quality bone.47-49

Consequently, if thigh pain is the cause of revision, bulky cobalt-chromium cementless implants should be avoided to reduce the risk of relapse.

Static pain, being due to distal overload, may be directly addressed by femoral shaft strengthening. The cortical strut grafting technique, originally proposed by Hedley and Firestone50 in 1993, increases diaphyseal rigidity by augmenting its cross-section area. The consequence would be a reduced mismatch with the implanted stem. Also a nervous mechanism has been suggested since the surgical exposure and the preparation of the lateral aspect of the diaphysis lead to extensive periosteal denervation.6

In 1997, Hedley and Firestone51 reported 12 good results out of 13 hips after 1- to 5-year follow-up. In 2000, Domb et al24 reported satisfactory pain relief in 6 of 7 cases. Although case series are few and small, the reported results are encouraging and, justify the use of this technique in pure static thigh pain unresponsive to nonoperative treatment.

Conclusion

The review of the main etiopathogenetic hypothesis about thigh pain shows an absolute dichotomy between the micromotions model and the overload model. Although clinical appearances of these conditions are similar, their causes, developmental mechanisms and possible remedies do not overlap.

This consideration induced us to distinguish dynamic pain, arising from micromotions of the tip, and static pain, determined by stress concentration around the tip.

References

  1. Mancuso CA, Salvati EA, Johanson NA, Peterson MGE, Charlson ME. Patient’s expectations and satisfaction with total hip arthroplasty. J Arthroplasty. 1997; 12:387-396.
  2. Robbins GM, Masri BA, Garbuz DS, Duncan CP. Evaluation of pain in patients with apparently solid fixed total hip arthroplasty component. J Am Acad Surg. 2002; 10(2):86-94.
  3. Bulow JU, Scheller G, Arnold P, Synatschke M, Jani L. Uncemented total hip replacement and thigh pain. Int Othop. 1996; 20(2):65-69.
  4. Campbell AC, Rorabeck CH, Bourne RB, Chess D, Nott L. Thigh pain after cementless hip arthroplasty. Annoyance or ill omen. J Bone Joint Surg Br. 1992; 74(1):63-66.
  5. D’Lima DD, Oishi CS, Petersilge WJ, Colwell CW Jr, Walker RH. 100 cemented versus 100 noncemented stems with comparison of 25 matched pairs. Clin Orthop Relat Res. 1998; (348):140-148.
  6. Brown TE, Larson B, Shen F, Moskal JT. Thigh pain after cementless total hip arthroplasty: evaluation and management. J Am Acad Orthop Surg. 2002; 10(6):385-392.
  7. Herzwurm PJ, Simpson SL, Duffin S, Oswald SG, Ebert FR. Thigh pain and total hip arthroplasty: scintigraphy with 2.5-year followup. Clin Orthop Relat Res. 1997; (336):156-161.
  8. Cameron HU, Bhimji S. Early clinical trials with a proximally-fixed uncemented hip stem. Contemp Orthop. 1998; 17:31-37.
  9. Bourne RB, Rorabeck CH, Ghazal ME, Lee MH. Pain in the thigh following total hip replacement with a porous-coated anatomic prosthesis for osteoarthrosis. A five-year follow-up study. J Bone Joint Surg Am. 1994; 76(10):1464-1470.
  10. Engh CA Jr, Culpepper WJ II, Engh CA. Long term results of use of the anatomic medullary locking prostesis in total hip arthroplasty. J Bone Joint Surg Am. 1997; 79(2):177-184.
  11. Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop Relat Res. 1990; (257):107-128.
  12. Lieberman JR, Huo MH, Schneider R, Salvati EA, Rodi S. Evaluation of painful hip arthroplasties: are technetium bone scans necessary? J Bone Joint Surg Br. 1993; 75(3):475-478.
  13. Costerton JW, Montanaro L, Arciola CR. Biofilm in implant infections: its production and regulation. Int J Artif Organs. 2005; 28(11):1062-1068.
  14. Engh CA, Bobyn JD, Glassman AH. Porous-coated hip replacement: the factors governing bone ingrowth, stress-shielding, and clinical results. J Bone Joint Surg Br. 1987; 69(1):45-55.
  15. Engh CA, Massin P. Cementless total hip arthroplasty using the anatomic medullary locking stem: results using a survivorship analysis. Clin Orthop Relat Res. 1989; (249):141-158.
  16. Engh CA, Bobyn JD. The influence of stem size and extent of porous coating on femoral bone resorption after primary cementless hip arthroplasty. Clin Orthop Relat Res. 1987; (231):7-28.
  17. Whiteside LA. The effect of stem fit on bone hypertrophy and pain relief in cementless total hip arthroplasty. Clin Orthop Relat Res. 1989; (247):138-147.
  18. Pipino F, Molfetta L, Grandizio M. Preservation of the femoral neck in hip arthroplasty: results of a 13- to 17-year follow-up. J Orthop Traumatol. 2000; 1:31-39.
  19. Pipino F. The bone-prosthesis interaction. J Orthop Traumatol. 2000; 1:3-9.
  20. Koch JC. The laws of bone architecture. Am J Anat. 1917; 21:177-298.
  21. Skinner HB, Curlin FJ. Decreased pain with lower flexural rigidity of uncemented femoral prosthesis. Orthopedics. 1990; 13(11):1223-1228.
  22. Bourkart BC, Bourne RB, Rorabeck CH, Kirk PG. Thigh pain in cementless total hip arthroplasty. A comparison of two systems at 2 years’ follow-up. Orthop Clin North Am. 1993; 24(4):645-653.
  23. Lavernia C, d’Apuzzo M, Hernandez V, Lee D. Thigh pain in total hip arthroplasty. The effects of elastic moduli. J Arthroplasty. 2004; 19(7):10-16.
  24. Domb B, Hostin E, Mont MA, Hungerford DS. Cotical strut grafting for enigmatic thigh pain following total hip arthroplasty. Orthopedics. 2000; 23(1):21-24.
  25. Fetto JF, Austin KS. A missing link in the evolution of THR: “discovery of the lateral femur”. Orthopedics. 1994; 17(4):347-351.
  26. Fetto J, Leali A, Moroz A. Evolution of the Koch model of the biomechanics of the hip: clinical perspective. J Orthop Sci. 2002; 7(6):724-730.
  27. Mach DB, Rogers SD, Sabino MC, Luger NM, et al. Origins of skeletal pain: sensory and sympathetic innervation of the mouse femur. Neuroscience. 2002; 113(1):155-166.
  28. Namba RS, Keyak JH, Kim AS, Vu LP, Skinner HB. Cementless implant composition and femoral stress: a finite element analysis. Clin Orthop Relat Res. 1998; (347):261-267.
  29. Wolff J. The Law of Bone Remodeling. Berlin, Heidelberg: Springer; 1986.
  30. Vresilovic EJ, Hozack WJ, Rothman RH. Incidence of thigh pain after uncemented total hip arthroplasty as a function of femoral stem size. J Arthroplasty. 1996; 11(3):304-311.
  31. Cameron HU. The two- to six-year results with a proximally modular noncemented total hip replacement used in hip revision. Clin Orthop Relat Res. 1994; (298):47-53.
  32. McAuley JP, Culpepper WJ, Engh CA. Total hip arthroplasty: concerns with extensively porous coated femoral components. Clin Orthop Relat Res. 1998; (355):182-188.
  33. Naumann T, Buchner F, Eckardt V. Pain in the thigh following Zweymuller’s total endoprosthesis: computerized tomography studies of the morphological behavior of the metadiaphyseal femur. Z Orthop Ihre Grenzgeb. 1987; 125(6):664-668.
  34. Schramm M, Keck F, Hohmann D, Pitto RP. Total hip arthroplasty using an uncemented femoral component with taper design: outcome at 10-year follow-up. Arch Orthop Trauma Surg. 2000; 120(7-8):407-412.
  35. Kim YH, Kim VE. Uncemented porous-coated anatomic total hip replacement: results at 6 years in a consecutive series. J Bone Joint Surg Br. 1993; 75(1):6-13.
  36. Rogers A, Kulkarni R, Downes EM. The ABG hydroxyapatite-coated hip prosthesis: one hundred consecutive operations with average 6-year follow-up. J Arthroplasty. 2003; 18(5):619-625.
  37. Gill TJ, Sledge JB, Orler R, Ganz R. Lateral insufficiency fractures of the femur caused by osteopenia and varus angulation: a complication of total hip arthroplasty. J Arthroplasty. 1999; 14(8):982-987.
  38. Estok DM II, Orr TE, Harris WH. Factors affecting cement strains near the tip of a cemented femoral component. J Arthroplasty. 1997; 12(1):40-48.
  39. Kristiansen B, Jensen JS. Biomechanical factors in loosening of the Stanmore hip. Acta Orthop Scand. 1985; 56(1):21-24.
  40. Panisello JJ, Herrero L, Herrera A, Canales V et al. Bone remodelling after total hip arthroplasty using an uncemented anatomic femoral stem: a three-year prospective study using bone densitometry. J Orthop Surg (Hong Kong). 2006; 14(1):32-37.
  41. Khalily C, Lester DK. Results of a tapered cementless femoral stem implanted in varus. J Arthroplasty. 2002; 17(4):463-466.
  42. Schneider U, Breusch SJ, Thomsen M, Wirtz DC, Lukoschek M. Influence of implant position of a hip prosthesis on alignment exemplified by the CLS shaft [in German]. Unfallchirurg. 2002; 105(1):31-35.
  43. Morrey BF, Adams RA, Kessler M. A conservative femoral replacement for total hip arthroplasty. A prospective study. J Bone Joint Surg Br. 2000; 82(7):952-958.
  44. Callaghan JJ, Dysart SH, Savory CG. The uncemented porous-coated anatomic total hip prosthesis. Two-year results of a prospective consecutive series. J Bone Joint Surg Am. 1988; 70(3):337-346.
  45. Barrack RL, Jasty M, Bragdon C, Haire T, Harris WH. Thigh pain despite bone ingrowth into uncemented femoral stems. J Bone Joint Surg Br. 1992; 74(4):507-510.
  46. Davis CM III, Berry DJ, Harmsen WS. Cemented revision of failed uncemented femoral components of total hip arthroplasty. J Bone Joint Surg Am. 2003; 85(7):1264-1269.
  47. Chandler HP, Ayres DK, Tan RC, Anderson LC, Varma AK. Revision total hip replacement using the S-ROM femoral component. Clin Orthop Relat Res. 1995; (319):130-140.
  48. Moreland JR, Bernstein ML. Femoral revision hip arthroplasty with uncemented, porous-coated stems. Clin Orthop Relat Res. 1995; (319):141-150.
  49. Paprosky Greidanus NV, Antoniou J. Minimum 10-year-results of extensively porous-coated stems in revision hip arthroplasty. Clin Orthop Relat Res. 1999; (369):230-242.
  50. Hedley AK, Firestone TS. Abstract: The use of an onlay cortical allograft strut for persistent thigh pain following cementless total hip replacement. Clin Orthop Relat Res. 1993; (295):172-178.
  51. Hedley AK, Firestone TS. Abstract: The use of an onlay cortical allograft strut for persistent thigh pain following cementless total hip replacement. Ortopaedic Transactions. 1997; 64:168.

Authors

Dr Pierannunzii is from Gaetano Pini Orthopaedic Institute, Milan, Italy.

Dr Pierannunzii has no relevant financial relationships to disclose. Dr Morgan, CME Editor, has disclosed the following relevant financial relationships: Stryker, speakers bureau; Smith & Nephew, speakers bureau, research grant recipient; AO International, speakers bureau, research grant recipient; Synthes, institutional support. Dr D’Ambrosia, Editor-in-Chief, has no relevant financial relationships to disclose. The staff of Orthopedics have no relevant financial relationships to disclose.

The material presented at or in any Vindico Medical Education continuing education activity does not necessarily reflect the views and opinions of Vindico Medical Education or Orthopedics. Neither Vindico Medical Education or Orthopedics, nor the faculty endorse or recommend any techniques, commercial products, or manufacturers. The faculty/authors may discuss the use of materials and/or products that have not yet been approved by the US Food and Drug Administration. All readers and continuing education participants should verify all information before treating patients or utilizing any product.

Correspondence should be addressed to: Luca M.C. Pierannunzii, MD, Via P. Finzi, 15-20126, Milan, Italy.

Sign up to receive

Journal E-contents