Introduction

Various etiologies account for severe alveolar ridge atrophies as non-restored edentulous spaces, (disuse atrophy), poorly constructed dentures, periodontitis and peri-implantitis [1, 2]. Jaws affected by neoplasms, trauma, and malformations require bone augmentation procedures to restore the function, anatomical features, and esthetics [3, 4]. The goal for prosthetic replacement is restoring the bone volume, through augmentation procedures utilizing harvested autogenous bone grafts,  allografts, xenografts, alloplasts, or combinations of materials [1]. Autogenous grafts are the gold standard, they combine the properties of osteoconduction, osteoinduction, and osteogenesis, without risk of infection transmission or  immunologic responses [5-7].

Iliac crest grafts are the most commonly utilized autogenous grafts for the ease of procedural accessibility and acquirement, and lowest morbidity [6, 8]. Iliac crest bone harvesting grafts, either anterior or posterior, depend on the type of surgery, as cancellous, cortico-cancellous, or vascularized grafts and allow for large quantities of bone available for harvesting [5]. Oral & maxillofacial surgeons prefer the anterior iliac crest due to the nature of their operations, requiring rich cancellous component needed for the augmentation, as well as  the patient’s position lying face-up [9, 10].

The aim of this study is to compare the pain and QoL and co-morbidity of patients who underwent anterior versus posterior iliac crest bone graft harvesting procedures at King Abdulaziz University Hospital (KAUH), at 3 different time intervals, and to correlate our findings with multiple subjective, objective, and socio-demographic factors.

Materials and Methods

A cross-sectional study was conducted using a post-operative survey instrument to obtain responses in a phone call interview. Confidentiality was secured through maintaining anonymous participants’ information. Demographic data including age, gender, time and site of operation were obtained. Clinical notes and medical records were reviewed. Sample selection included all patients treated by the senior investigator (MYA) at KAUH and were under his care postoperatively from the period of Jan 1, 2012, to Aug 31, 2020. 31 patients were included (15 females and 16 males). 7 patients underwent posterior iliac crest, while 24 had their grafts harvested from the anterior iliac crest.

Questionnaire design

A single investigator (MRA) contacted all participants and informed them about the questionnaire thoroughly. No participants faced any difficulty answering the survey.  A phone call-based interview was conducted three days later to obtain their verbal consent and the answers to the survey questions.  (MRA) recorded all the answers on a Google Sheet document (Google Sheets, Google LLC, California), which was exported later to Microsoft Excel (Microsoft Excel, Microsoft Corporation, Washington DC) for data analysis purposes.

The questionnaire consisted of 3 main parts. The first part included general demographic questions (Table 2). The second part consisted of only the first section of the Harris Hip Score [11], and since the entire sample was of Muslim religion, 2 questions regarding prayer were added: (were you able to perform Prayer? yes, normally; on a chair; or lying down. If the previous answer was “on a chair,” did you pray on a chair before the procedure?. Arabic translation of the Harris Hip Score was validated for reliability using Cronbach’s alpha.  The third part consisted of post-operative co-morbidities (Table 1). The questionnaire was validated for both content and face validity by the research team and three oral and maxillofacial surgery consultants.

Table 1. Post-operative co-morbidities

Surgical technique for the anterior iliac approach

The incision was carried out through the skin, subcutaneous layer, Camper’s fascia, Scarpa’s fascia, and ends on the periosteum. The lateral aspect was minimally dissected preserving the tensor fascia lata. Encroachment on the anterior-superior iliac spine was avoided to prevent injury to the inguinal ligament or the other six attachments.  The medial aspect was fully retracted using Taylor retractor. 

Two parallel anterior and posterior cuts on the medial aspect were connected via a superior crestal cut using a surgical saw, followed by a medial inferior osteotomy stop cut with a round bur.

Subsequently, serial osteotomes were used securing that their angles  avoid harvesting a bicortical graft and or perforate the lateral cortical aspect of the bone.

Finally, cortex was delivered from the anterior aspect, measuring 2 by 4 centimeters, and excavating the amount of the needed cancellous bone,  around 50CC, including the hip block from the anterior iliac crest. 

The site was then irrigated, and a hemostatic agent (usually microfibrillar collagen sheets) placed on the medial aspect, the wound was closed  in layers,  followed by sterostrip application and Tensoplast or Elastoplast on top of the surgical site. 

The surgical technique for the posterior iliac approach

Landmarking started from the posterior superior iliac spine, and a curvilinear mark was marked along the superior aspect of the gluteus maximus muscle.

A curvilinear incision was done around that line, measuring  7 to 8 centimeters, avoiding  injury of  S1, S2, S3 or L1, L2, L3 branches. 

Dissection through four layers, skin, subcutaneous layer, lumbodorsal fascia, ended on the periosteum, exposing the bone, and avoiding  sciatic nerve exposure. Osteotomy was carried out through the surgical saw approach. 

Anterior and posterior trapezoidal-like cuts connected from the crestal aspect, ended by inferior osteotomy stop cut with a round bur, followed by serial osteotomized cuts.  Collected, bone block was, a 5-by 4-centimeter. Cancellous bone was excavated from the site, which along with the block, accounted for 70 to 80 and even up to 100 CCs in total, followed by irrigation and application of a hemostatic agent, mainly microfibrillar collagen.  Insertion of a hemovac or a Jackson Pratt drain, size 16, with an exit through the skin, was followed by  periosteum closure, and layers with  application of sterostrip and pressure dressing.

Pain management

During the admission period, the patients were placed on narcotics, mainly morphine, 2 to 4 milligrams IV, with NSAIDs and paracetamol to wean the narcotic requirement before discharge. The NSAIDs used in the intraoperative course were between Toradol® (Roche Holding AG, Basel, Switzerland) or  ibuprofen, which can go along with Perfalgan® (Aspen Laboratories Pvt Ltd, Delhi, India). 

The at-home course was the combination of Solpadine (Omega Pharma, Nazareth, Belgium) and ibuprofen, usually two tablets of Solpadine 500mg and one tablet of ibuprofen 600 or 400 mg, depending on the patient’s weight, with daily water intake, 2.7 liters  for the females or 3.7 liters for the males. 

Statistical analysis

Data analysis used IBM SPSS version 23 (IBM Corp., Armonk, NY, USA). A simple descriptive statistic was used for categorical and nominal variables, while continuous variables were presented by mean and standard deviations. An average score was calculated called “QoL” derived from the questions with a hundred-point scale as follows:

• Pain 
  • None, or ignores it=100 
  • Slight, occasional, no compromise in activity=80 
  • Mild pain, no effect on average activities, rarely moderate pain with unusual activity, may take aspirin=60
  • Moderate pain, tolerable but makes concessions to pain=Some limitations of the ordinary activity or work=40
  • Marked pain, serious limitation of activities=20 
  • Totally disabled, crippled, pain in bed, bedridden=0

• Support 
  • None=100
  • Cane/Walking stick for long walks=80
  • Cane/Walking stick most of the time=60
  • One crutch=40
  • Two Canes/Walking sticks=20
  • Two crutches or not being able to walk=0

• Distance walked 
  • Unlimited=100
  • Six blocks (30 minutes)=75
  • Two or three blocks (10-15 minutes)=50
  • Indoors only=25
  • Bed and chair only=0

• Limp 
  • None=100
  • Slight=66.7
  • Moderate=33.3
  • Severe or unable to walk=0

• Activities- shoes, socks 
  • With ease=100
  • With difficulty=50
  • Unable to fit or tie=0

• Using stairs 
  • Normally without using a railing=100
  • Normally using a railing=50
  • Unable to do stairs=0

• Sitting 
  • Comfortably, an ordinary chair for one hour=100
  • On a highchair for 30 minutes=50
  • Unable to sit comfortably on any chair=0

• Were you able to drive/leave the house? 
  • Yes=100
  • No=0

• Were you able to perform Salah (Prayer)? 
  • Yes, normally=100
  • Yes, on a chair=50
  • Yes, lying down=0

Each question is separated by the time interval of IAS, 6w, and Current Time. A Reliability Analysis was used with an Alpha (Cronbach) model to study the properties of measurement of the “QoL” scale, the items that compose it, and the average inter-item correlation. A Pain scale was also used in the comparison by converting the existing scoring of “Rate the pain from 0-10” by multiplying each response by tens to achieve a hundred-point equivalent score for each time interval of IAS, 6w, and Current Time. After conversion and calculation of the scales, a repeated measure ANOVA with a Greenhouse-Geisser correction was used as a test to correlate the scales to the demographical data. Lastly, a conventional p-value <0.05 was the criteria to reject the null hypothesis.

Results and Discussion

The QoL and pain levels at different time points were investigated. Most of the patients underwent anterior iliac crest graft harvesting (77%, n=24). Other demographic and characteristics of the participants are shown in Table 2.

Using a 100-point scale, the mean pain scores showed decrease from the IAS period (63.23 ± 30.4) to after 6w (30.97 ± 30.4) and to current time (4.84 ± 11.2), suggesting an improvement in their pain management, a similar trend was found in the other evaluated aspects (Figure 1).

The pain profile showed a decreasing trend of mean pain scores among patients across time, (IAS), after 6w, and at the current time, with improvement between IAS and 6w (93.5%, n=29) and between 6w and the current time (96.8%, n=30).

Cronbach’s alpha was used to test the reliability (internal consistency) of the items composing the QoL scale at different time points. The values obtained were 0.893 for IAS, 0.885 for 6w, and 0.678 for the present. Using the rule of thumb which states that > 0.9 is excellent, > 0.8 is good, > 0.7 acceptable, > 0.6 – Questionable, > 0.5 – Poor, and < 0.5 is unacceptable”, items in QoL at IAS and 6w had a good internal consistency, while QoL at current time was questionable, due to  variation of  time elapsed since surgery differs amongst the patients.

Table 3 demonstrates multivariate tests on the QoL against time and other socio-demographic factors. Results revealed a significant effect between the QoL and time alone (p=0.002). ANOVA with repeated measures with a Greenhouse-Geisser correction showed mean QoL scores to be statistically significantly different between time points (F(2.000, 23.000) =8.260, p=0.002). Using Mauchly’s Test of sphericity, a p-value of 0.806 was obtained, suggesting normality in the distribution of the statistical data. The normality in the distribution of patients was explicitly found in the QoL at IAS, at 6w, and at current periods, by p-values greater than 0.05 obtained using Levene’s Test of equality of error variances.

Estimated marginal means were measuring the improvement in the QoL of patients at different time points. The estimated mean QoL increased, as shown by values of 43.801 (SE=3.401) at IAS, 76.709 (SE=4.106) at 6w, and 94.177 (SE=1.621) at the current time. The pairwise comparison revealed that the patients had an overall statistically significant improvement (p < 0.001) in the QoL at all three time points (Figure 2).

Table 2. Demographic characteristics, pain, and QoL scores of 31 study samples

Multivariate tests on the pain scores against time and other socio-demographic factors were performed (Table 4). Results revealed a significant effect between the pain level and time alone (p=0.006) and between pain level and time interacting with age (p=0.006), Moreover, ANOVA with repeated measures with a Greenhouse-Geisser correction revealed that the mean pain scores were statistically significantly different between time points [(F(2.000,23.000)=6.413, p=0.006), (F(2.000,23.000)= 6.462, p=0.006)]. Using Mauchly’s Test of sphericity, a p-value of 0.304 was obtained, suggesting lower normality in the distribution of the statistical data, less normality in the distribution of patients was still found specifically in the pain scores in the IAS, 6w, and at the current time groups, as shown by p-values less than or equal to 0.05 obtained using Levene’s Test of equality of error variances.

Table 3. Multivariate Tests on QoL against time and other socio-demographic factors.

The estimated marginal means were measured depicting improvement in the pain levels experienced at different time points (Figure 3). The pairwise comparison revealed that the patients had an overall statistically significant improvement (p < 0.001) in the pain scores in the IAS, 6w , and at the current time.

The QoL score means for each patient in both the anterior and posterior approach groups, respectively, along with their relevant demographic data, are presented in all time intervals in Tables 5 and 6.

Table 4. Multivariate Tests on pain scores against time and other socio-demographic factors.

Regarding the specific donor site, the majority of the patients with anterior iliac crest harvest rated their pain IAS to be 7 or above out of 10 (54.2%, n=13), while 37.5% (n=9) needed support to move around IAS. 29.2% (n = 7) stopped support needs 6w after surgery, while 8.4% still needed support until the current time, with the mean of the current time being 3.17±1.7 years.

62.5% (n=15) of the patients had limitations in the walking distance IAS, with good ability to move around indoors (37.5%, n=9). While, 54.2% (n=13) could walk unlimited distance after 6w,, and 87.5% (n=21) in the current time. Regarding limping, 66.7% (n=16) of the patients experienced limping IAS, which decreased to 25% (n=6) at 6w, and none at the current time. 66.7% (n=16) of the patients faced difficulties in performing daily activities IAS, while only 29.2% (n=7) faced difficulty at 6w. 33.3% (n=8) were unable to use the stairs IAS, but only 2 (8.3%) were still unable to use the stairs at 6w, but all patients can use the stairs normally at the current time. Most patients (62.5%, n=15) could not sit comfortably on any chair IAS. Most of the patients (66.7%, n=16) could not leave the house IAS due to pain (54.2%, n=13), tiredness and fatigue (8.3%, n=2), or fear of contaminating the wound (4.2%, n=1). Post-operative numbness was reported over the scar (25%, n=6), in the hip (20.8%, n=5), or in the leg (12.5%, n=3), 83.3% were able to live with it. The majority of  anterior iliac crest graft (54.2%, n=13) were satisfied with their scar, while 29.2% (n=7) were indifferent. For immediate postoperative complications, only 12.5% (n=3) complained of swelling, 20.9% (n=5) hematoma and only 8.3% (n=2) had procedural site inflammation.

Table 5. QoL score means for anterior approach patients

Table 6. QoL score means for posterior approach patients