Oral antibiotics alone were shown in some trials to cure the majority of diabetic foot infections [42,50]. A 2001 report by Senneville et al. on oral rifampin plus ofloxacin for a median duration of six months achieved an 88.2% cure, defined by the disappearance of all signs and symptoms of infection and the absence of relapse during follow up [51]. In a composite review by Conterno et al. (2013), there was no significant difference either in the rates of remission after treatment with oral versus IV antibiotics in chronic osteomyelitis caused by sensitive pathogens, or in the rates of mild, moderate, or severe adverse events between the two groups [7]. Similarly, a 2012 systematic review of studies on osteomyelitis from 1970 to 2011 revealed that the success rates were similar for both routes of administration [1].

A multicenter, open-label, parallel-group, randomized, controlled noninferiority trial, Oral versus IV Antibiotics for Bone and Joint Infections (OVIVA) [15], evaluated outcomes at one year of IV versus oral antibiotics for the first six weeks of treatment in 1054 patients. This trial challenged the widely accepted standard of care and concluded that oral antibiotic therapy was noninferior to IV therapy when used during the first six weeks for complex orthopedic infection. Complications were more common in the IV group who also had a considerably longer median hospital stay with no significant difference in the incidence of Clostridioides difficile-associated diarrhea, the percentage of participants reporting at least one serious adverse event, or the treatment failure at one year (13.2% in the oral group vs. 14.6% in the IV group). Since its publication, the OVIVA protocol has been implemented in some clinical practices. A study from a British orthopedic hospital showed that patients who were switched to an oral regimen post OVIVA-implementation had a shorter length of stay and reduced cost of care, without a significant difference in clinical outcomes [49]. Highly bioavailable agents with good bone penetration and biofilm activity, such as rifampin and ciprofloxacin, were more commonly used in the oral subgroups. Definitive treatment failure was more common following implementation (13.6% vs 18.6%), although there was no statistical difference between the infection-free survival curves at 12 months, which was similar to those seen in the OVIVA trial [49]. Another study from the Veterans Administration Medical Center in Iowa City implemented a quality improvement protocol aimed at decreasing outpatient parenteral antimicrobial therapy (OPAT) and increasing oral antibiotic use. The outcomes demonstrated significantly lower lengths of stay and no difference in the 6-month recurrence rates or mortality [52]. However, aside from OVIVA, no other large randomized controlled trials to date have determined whether enteral antibiotic therapies are non-inferior to parenteral antibiotic therapies for the empiric treatment of osteomyelitis. Other smaller studies demonstrated the non-inferiority of enteral compared to parenteral therapy for bone and joint infections. Using oral therapies as opposed to IV has significant improvement potential, including reduced utilization of IV lines [53], shorter length of stay, lower cost [49,54], and decreased adverse effect profiles, including nephrotoxicity [55].

Diabetes mellitus, according to the National Institute of Diabetes and Digestive and Kidney Diseases, affects 37.3 million people in the United States—approximately 11.3% of the population in 2019 [56]. The most common indications for hospital admission in diabetics are soft tissue and bone infection of the lower limbs [57]. About one-third of diabetics with foot infections have osteomyelitis [42], the most common cause for amputation in the infected diabetic foot. According to the 2020 National Diabetes Statistics Report, 130,000 adults had lower extremity amputations relating to diabetes (5.6 per 1000 adults with diabetes) [58]. Following an amputation, about one-third of patients will undergo an amputation of their other limb within three years, and two-thirds will die in five years [59]. The prompt diagnosis and appropriate treatment of diabetic foot osteomyelitis may help prevent amputation, with its psychological, social, and financial consequences, reduce morbidity and mortality, as well as decrease the burden on the healthcare system.

Emerging evidence reveals that most infections respond well to antibiotics alone [60]. The early excision of all infected bone is deemed necessary by some authors, only necrotic bone removal is suggested by others, and limited debridement in the clinic—with surgery restricted to patients who are unresponsive to antibiotics—is suggested by others still [61]. The conventional advice that that excision of infected and necrotic bone along with aggressive parenteral therapy is crucial in the management of diabetic foot osteomyelitis was challenged by a 10-year retrospective review [46]. The majority of the 22 patients identified to have overt osteomyelitis were successfully managed with prolonged courses (12 weeks median) of oral antibiotics (most commonly, clindamycin 600–1800 mg daily) as outpatients with limited debridement undertaken in the clinic, without the need for hospitalizations for either debridement or parenteral antibiotics. Data from another study [61] demonstrated similar results, suggesting that conservative management with antibiotics alone, whether oral or IV, can be successful in the majority of cases, except when surgery is clearly indicated. An analysis published in JAMA in 1995 concluded that a 10-week culture-guided post-surgical debridement oral antibiotic therapy, in patients without systemic infection, may be as effective and less costly than other approaches [62]. A retrospective medical record review of 325 diabetic patients with foot osteomyelitis revealed successful treatment with oral antibiotic therapy, with or without debridement, in almost 80% of cases, with a mean duration of therapy of 40 ± 30 weeks. This study also concluded that acceptable results from oral therapy may be attained even with little operative facilities and resources [63].

While mild and non-limb-threatening infections are generally monomicrobial and can thus be treated successfully with a single antibiotic, severe and/or limb-threatening infections are usually polymicrobial, involving both aerobic and anaerobic organisms [45,64,65,66]. Gram-positive cocci including Staphylococcus aureus, coagulase-negative staphylococci, group B streptococci, Enterococci, and Corynebacterium species are the most common cultured organisms, but gram-negative bacilli and/or anaerobes may frequently be encountered as well. Accordingly, empiric coverage targeting gram positives and gram negatives, as well as aerobes and anaerobes is recommended in most situations. A prospective, randomized, multicenter trial comparing the efficacy of two broad-spectrum regimens initially administered intravenously (ofloxacin vs ampicillin/sulbactam) then orally (ofloxacin vs amoxicillin/clavulanate) to two treatment groups with similar infection severity showed that 67% of the pathogens were gram-positive cocci and 92% were aerobic organisms. Both regimens proved to be effective after a total duration of about three weeks [67]. In some settings, however, empiric therapy needs to be different from the published guidelines, depending on the local prevalence of microorganisms. The results of a multicenter descriptive and analytic cross-sectional study from 17 centers of four Latin American countries found that most infections, unlike in other continents, were monomicrobial, and gram negatives had a high prevalence in mild diabetic foot infections. A combination empiric treatment, amoxicillin-clavulanate with trimethoprim-sulfamethoxazole, was suggested for diabetic foot infection if osteomyelitis is probable [68]. In Europe, rifampin is used as adjunctive therapy to the backbone oral antimicrobial treatment for osteomyelitis, including diabetic foot, and the results suggest that it may improve amputation-free survival [69]. While mild and moderate infections may be treated with oral antibiotics alone, oral therapy may not be appropriate for patients with systemic illness, severe infection, poor enteral absorption, vasculopathy, or infections caused by organisms resistant to oral agents. Additionally, with the prevalence of comorbid peripheral artery disease in diabetic foot infections, diabetic patients with foot osteomyelitis should be evaluated for limb ischemia and undergo revascularization as appropriate [70]. With concomitant revascularization, patients may be successfully treated with medical therapy and avoid amputation [8]. Until proper blood supply is established, diabetes is a risk factor for treatment failure in chronic osteomyelitis treated with prolonged suppressive oral antibiotics [3]. A few case reports describe the successful treatment of osteomyelitis of the proximal and distal phalanx of the toes with IV antibiotics followed by 10 weeks of oral antibiotics after successful angioplasty. Optimal outcomes in those reports are defined as resolution of the ulcers, radiographic defects, and complete restoration of foot function [8]. The paucity of similar case reports renders such an observation less robustly generalizable. The treatment of osteomyelitis using orally administered antibiotics is also favorable in the high-risk niche persons who inject drugs (PWID). A retrospective analysis by Marks et al. showed that in PWID with invasive bacterial infections who left against medical advice, prescribing no oral antibiotics at discharge compared to oral antibiotics, was associated with 2.32 times higher odds of 90-day readmission. They also noted that the 90-day admission rate was similar in people who were discharged on oral antibiotics compared to those had a full IV antibiotic course [71], another indirect but strong indication that IV and oral therapies are equally efficacious in this subpopulation of patients.

Oral antibiotics for the treatment of osteomyelitis in specific bones have also been described, primarily through case studies and smaller series. For example, in jawbone and joint infections, common pathogens include viridans Streptococci, mixed anaerobic flora, coagulase-negative Staphylococcus, Actinomyces, Eikenella corrodens, Candida species, Neisseria species, Enterobacteriaceae, and Staphylococcus aureus [17]. The retrospective analysis of jaw osteomyelitis by Lim et al. showed better outcomes at two months posttreatment in patients treated with oral antibiotics after surgery compared to those treated with IV antibiotics. The oral antibiotics that have been successfully used for jaw osteomyelitis treatment include amoxicillin, amoxicillin/clavulanic acid, clindamycin, moxifloxacin, cotrimoxazole, and penicillin [17]. As for bone and joint infections in the hand, a retrospective study of 61 patients with acute inoculation osteomyelitis of the hand were treated with oral antibiotics for six months; this resulted in a 100% cure rate [11]. The antibiotics used included trimethoprim/sulfamethoxazole, levofloxacin, and moxifloxacin. Common pathogens were methicillin-resistant Staphylococcus aureus (MRSA) (22%), Staphylococcus epidermidis (18%), methicillin-resistant Staphylococcus aureus (MSSA) (13%), and Streptococcus species (10%).

Staphylococcus aureus is a leading cause of bone and joint infections (BJI), culture proven in as many as 75% of cases. Staphylococcus species have evolved over the decades to acquire resistance to beta-lactam antibiotics, such that Staphylococcus aureus is distinguished as being methicillin sensitive or methicillin resistant. More than one-third of all staphylococcal BJIs are caused by MRSA, with an increasing prevalence of methicillin resistance [72]. Physicians have historically treated BJI with empiric parenteral antibiotics that are effective against MRSA, and de-escalated only if MRSA is ruled out based on culture and/or molecular testing data. This practice came about when many of the currently available anti-MRSA enteral antibiotics did not yet exist. The recent increasing availability of efficacious and highly bioavailable oral antibiotics in the management of MRSA infections has challenged this historical practice [73]. One important gap in the knowledge is whether multi-drug enteral regimens are superior to monotherapies because it has long been known that causative organisms of BJIs create biofilms that act as a mechanical barrier and prevent the delivery of the antibiotic to the nidus of infection. Some antibiotics, like rifampin, penetrate these biofilms and work synergistically with other agents to overcome this bacterial defense mechanism [74].

For the initial treatment of chronic osteomyelitis caused by MSSA, parenteral beta-lactam agents, i.e., oxacillin, nafcillin, and cefazolin, are suitable options. Unfortunately, nafcillin and cefazolin are not available in an oral formulation, and the bioavailability of oral penicillin, oral formulations of oxacillin, and cephalosporins, is usually low. Therefore, switching to oral therapy often requires a change to other agents active against both MSSA and MRSA, such as doxycycline, clindamycin, linezolid, and/or trimethoprim-sulfamethoxazole. Rifampin should be combined with other agents, namely fluoroquinolones or linezolid [2]. Hence, rifampin has a niche as a “biofilm active agent”. It is best studied for staphylococcal prosthetic joint infection in the setting of hardware retention [75,76,77]. Data were accordingly extrapolated for other hardware infections such as osteofixation and spinal implant. Its use results in lower treatment failures in prosthetic joint infections (PJIs) with implant retention. However, it has significant pharmacologic interactions (primarily via the CYP 3A4 pathway) that need to be considered prior to its clinical utilization. Published studies recommend against rifampin use in combination with fusidic acid [78,79], the use of which is uncommon in the United States. Another caution is co-prescribing oral clindamycin and rifampin, as clindamycin concentrations can be substantially decreased due to increased first pass metabolism resulting from P450 enzyme induction by rifampin [80]. However, outside of those cautions, the use of rifampin is an important adjunct in the management of staphylococcal osteomyelitis, particularly those involving biofilms, such as with prosthetic implants. Discouraged rifampin usage includes monotherapy (due to prompt risk of resistance emergence) [81] and prior to surgical debridement without a partnered antimicrobial. When choosing between oral antibiotics, a retrospective analysis by Nguyen et al. found the rifampin-trimethoprim-sulfamethoxazole combination to be as effective as rifampin-linezolid, with trimethoprim-sulfamethoxazole being cheaper than linezolid [16].

Given their favorable PK/PD data, fluoroquinolones have been well studied and have shown great efficacy in the management of osteomyelitis. They are bactericidal against most gram-negative aerobic bacilli, with ciprofloxacin having specific indications for the treatment of Pseudomonas aeruginosa. Previous literature has shown that quinolones are successful in treating pseudomonal and other gram negatives, in addition to other organisms, such as Staphylococcus aureus [83,84,85]. Papers published in the early 1990s highly recommended quinolones. They demonstrated success for the treatment of pseudomonal and other gram negatives, in addition to other organisms such as Staphylococcus aureus. Most of those references touted the favorable adverse effect profile of fluoroquinolones, with almost no reports of tendon damage at that time [86,87,88]. Those recommendations have changed over time—now fluoroquinolones are generally considered effective second line agents for infections caused by sensitive organisms [89]. Some of those early papers cautioned that the indiscriminate use of fluoroquinolones carries inherent concerns of potential failure in the treatment of staphylococcal osteomyelitis [40], over time falling completely out of favor for the treatment of staphylococcal infections. Of the quinolones, ciprofloxacin has been studied the most [83,86,90]. In 1988, the Swedish Study Group reviewed the use of oral ciprofloxacin in the management of gram-negative osteomyelitis [83]. This was an open, non-comparative, multi-center trial in 17 Swedish hospitals that evaluated a total of 34 patients with osteomyelitis in different bones. Patients were treated with various doses of ciprofloxacin without any parenteral therapy. There was resolution in 22 (65%) patients, improvement in 5 (14%), and failure in 7 (21%); thus ciprofloxacin was recommended as an oral alternative to IV antibiotics for the treatment of acute or chronic osteomyelitis caused by sensitive gram-negative bacilli including Pseudomonas sp. [88]. In 1990, Gentry et al. demonstrated the safety and efficacy of oral ciprofloxacin compared to IV therapy for infections caused by a wide variety of organisms, with a high success rate of 77% and no significant adverse events [86].

Oral ciprofloxacin in doses of 750 mg twice daily for the treatment of chronic osteomyelitis was assessed in hospitalized adult patients who presented with chronic osteomyelitis in bone biopsy and an organism susceptible to ciprofloxacin. When Enterococcus faecalis was isolated, ampicillin was added after ensuring that it had no activity similar to the test drug against the Gram-negative organism. The duration of treatment ranged from 28 to 254 days. The oral treatment with ciprofloxacin proved to be useful for the prolonged therapy of chronic osteomyelitis, always combined with surgical debridement. Oral therapy allows for easy outpatient use, good tolerability, and is effective; however, it demands special attention for the possible emergence of resistance, particularly in Staphylococcus aureus. It is important to note that 11 patients (65%) had been treated with other antibiotics before admission to the study [30].

There is paucity of evidence in the literature for the definitive treatment recommendations of unusual organisms. Reliance on the antibiogram and “standard” antimicrobial regimens for those organisms is what guides the choice of treatment.

Infections caused by Kingella kingae typically occur in patients with immune-compromising conditions. This organism is classically associated with endocarditis, bacteremia, and spondylodiscitis in adults. In a case report by Wilmes et al, high-dose oral amoxicillin therapy was used for pubic osteomyelitis, without surgical debridement, with good clinical and radiographic response to treatment after three months [22].

Osteomyelitis and septic arthritis caused by Yersinia enterocolitica is rare. Its most common portal of entry is the GI tract following the ingestion of contaminated food, water, or milk. Isolates of serotypes O3, O9, and O8 are the most frequent causes of sporadic human disease worldwide. This organism may cause infections in individuals without major underlying disease or specific risk factors. In a study where oral ciprofloxacin therapy was used with therapeutic success. The drug of choice is yet to be identified [29].

Discitis and vertebral osteomyelitis caused by Fusobacterium nucleatum have been treated with IV ertapenem for eight weeks initially together with oral amoxicillin/clavulanate for a total of 10 weeks. At 1-month follow-up after the completion of treatment, the patient’s inflammatory markers returned to normal values, and the infection resolved with L3–L4 auto-fusion [23].

Group G streptococcal osteomyelitis is rare, with fewer than 15 cases reported in the literature. A case report of a 71-year-old otherwise healthy male with osteomyelitis of the proximal femur was treated with IV penicillin for six weeks, followed by oral cephalexin for six months with a good outcome. The optimal dosage and duration of antibiotic therapy for group G streptococcal osteomyelitis, as well as the role of surgical debridement, are controversial [94].

An extremely rare case of Salmonella potsdam vertebrae osteomyelitis was confirmed via tissue culture and abscess fluid obtained during surgery. Based on a drug-sensitivity test, levofloxacin and ceftazidime were administrated through IV injection for three weeks, followed by oral antimicrobial therapy for another three weeks. At the 4-month follow up, back pain had almost completely resolved; the patient’s MRI demonstrated an improvement of swelling, with noted radiographic changes of edge sclerosis and L4/5 partial fusion [95].

Aggregatibacter actinomycetemcomitans is well known as the pathogen behind gingivitis and periodontitis. Discitis and vertebral osteomyelitis cases caused by this organism have rarely been reported. A successful antimicrobial therapeutic strategy for discitis with this organism is ampicillin or amoxicillin, but no cases have been reported using levofloxacin. Uno et al. reported a case where levofloxacin was selected due to unclear susceptibility results to amoxicillin (the organism failed to grow). It was used after two weeks of IV ceftriaxone. It was originally prescribed for two weeks, yet it was extended to six weeks due to elevated CRP and exacerbated low back pain two weeks post discharge. Based on that report, levofloxacin, to which A. actinomycetemcomitans is usually susceptible, can be an effective alternative oral antimicrobial agent when amoxicillin or ampicillin cannot be utilized. Six-week parenteral or highly bioavailable oral treatment is recommended in the case of discitis or vertebral osteomyelitis. This organism is usually susceptible to cephalosporins, rifampin, tetracyclines, or fluoroquinolones, and in vitro susceptibility to penicillin and ampicillin is variable. However, the clinical efficacy of fluoroquinolone therapy for this organism cannot be generalized based on this report, but it should be considered if no other options are available [36].

Malincarne et al. in an open three-armed non-randomized trial, evaluated moxifloxacin penetration into bone to evaluate its potential role in the treatment of bone infections. They determined plasma and bone moxifloxacin concentrations following the oral administration of single or double doses (400 mg every 12 h) [96]. The recovered plasma and bone concentrations after single administrations showed a stable bone/plasma ratio without a relevant reduction in plasma or tissue drug levels. Considering an MIC90 of 0.12 mg/L for methicillin-susceptible staphylococci and of less than 1 mg/L for most Enterobacteriaceae, the recovered mean moxifloxacin concentrations show that single dosing leads to bone and plasma moxifloxacin levels exceeding the MICs for the most relevant pathogens. Double moxifloxacin administration gives significantly higher plasma and bone concentrations, with an average of above 2.5 mg/L both in cancellous and cortical bone. This value exceeds the MIC90 of moxifloxacin for methicillin-resistant staphylococci, as reported by most authors, of 2 mg/L, and the clinical M. tuberculosis susceptibility breakpoint for moxifloxacin of 1 mg/L. The results demonstrate a good degree of penetration of moxifloxacin into bone [96].

Despite concerns about its bone penetration and poor bioavailability [97], amoxicillin/clavulanic acid, with its beta-lactam/beta-lactamase combination among the most frequently prescribed oral beta-lactam antibiotics worldwide for diabetic foot infections. It is considered the drug of choice by some, with activity against MSSA, streptococci, enterococci, many gram-negative rods, and anaerobes [98,99]. A retrospective cohort analysis among 794 diabetic foot infection episodes—including 339 diabetic foot osteomyelitis cases— found that oral amoxicillin/clavulanic acid resulted in similar clinical outcomes to other regimens and was a reasonable option when treating diabetic foot infections and osteomyelitis [100]. Other antibacterials used orally with successful outcomes are clindamycin (especially in patients with penicillin allergy) and fluoroquinolones (for gram-negative infections or in combination with amoxicillin/clavulanic acid or clindamycin). For limb-threatening infections, clindamycin, an aminoglycoside, and ampicillin as a triple antibacterial regimen have been found to be successful.

Multiple cost analysis studies have demonstrated the superiority of oral antibiotics for osteomyelitis. In a retrospective study by Bhagat et al. 73% of patients receiving outpatient IV antibiotics (OPAT) were found to be candidates for oral antibiotics per OVIVA criteria, and substituting oral for IV antibiotics could have resulted in an estimated average savings per patient of USD 3270.69 [10]. A similar analysis by Marks et al. conducted in the United Kingdom, found that 79.7% of patients were eligible for oral antibiotics, with an estimated savings of GBP 2950 (USD 3605) [12]. A scenario analysis for early discharge and outpatient oral treatment for osteomyelitis for the National Health Systems (NHSs) in Italy, Greece, and Spain suggested a positive impact in terms of the incidence of hospital-acquired infections, hospital bed saving/increased productivity, and reduced direct health care costs [13]. In their study of patients being treated for osteomyelitis of the hand with oral antibiotics, Henry et al. calculated a differential direct cost savings of 98% when compared to IV therapy [11]. In their systematic analysis of the 25 prospective studies comparing IV-only therapy to IV followed by oral stepdown, Wald-Dicker et al. also reported prolonged inpatient hospitalization in the IV-only group [6], which would translate to increased healthcare utilization costs.