Inhaled antibiotics are a cornerstone of treatment for various diseases such as cystic fibrosis (CF), non-CF bronchiectasis, and ventilator-associated pneumonia. The variety of available antibiotics is vast, and clinicians typically select them based on the local prevalence of microorganisms, specific antibiograms, antibiotic availability in their hospital, and patient-specific factors such as tolerability test results or administration preferences. The advantages of inhaled antibiotic administration include delivering higher drug concentrations directly to the infection site, avoiding the systemic side effects associated with parenteral or oral antibiotics, rapid action, and bypassing the liver's first-pass metabolism. Their therapeutic potential is considerable.1

However, inhaled antibiotics are not free from side effects. The most commonly reported adverse events are cough, tachycardia, hyper- or hypotension, and hypoxemia.2,3 Bronchospasm, often thought to be a common side effect of nebulization, is infrequently reported.3 To minimize these adverse effects, inhaled antibiotic solutions should be non-pyrogenic, sterile, preservative-free, with a pH level between 4 and 8, and an ideal osmolality between 150 and 550 mOsm/kg.4 Some experts contend that osmolality up to 1100 mOsm/kg can be used,5 but most authors agree that side effects increase as osmolality rises.6,7 In addition, tolerability is improved when solutions contain at least 30 mEq/L of permeant anions.7 Most researchers agree that chloride is the preferred anion, and aerosols with <30 mEq/L of chloride induce coughing.7 The ideal concentration is approximately 70 mEq/L of chloride.6

Although some studies suggest that dry powder inhalers may have better outcomes than nebulized solutions (such as time to first exacerbation, quality of life, or pharmacokinetics8,9), their tolerability may often be worse.10 Moreover, dry powder formulations for many antibiotics are not always commercially available, and when they are, they are generally more expensive than diluted formulations. Regarding nebulized solutions, there are three categories: (1) antibiotics approved for inhalation and commercialized as ready-to-use liquids (e.g., tobramycin 300 mg/5 mL), (2) antibiotics indicated for inhalation but requiring patient manipulation, such as powder reconstitution (e.g., colistimethate or aztreonam lysine), and (3) off-label use of intravenous antibiotics for inhalation, where patients must reconstitute (e.g., ampicillin) and dilute (e.g., gentamicin). In the latter two cases, and particularly in the third, there is little consensus on how to reconstitute and dilute antibiotics, especially regarding dose and diluent, and only 50% of studies report how the mixtures are prepared.11 Furthermore, fewer than 10% report data on osmolality, pH, or ion concentrations, which directly affect tolerability.11

If possible, intravenous solutions should be avoided as they may contain particles that induce inflammation and bronchospasm upon contact with the bronchial epithelium.12 Additionally, intravenous formulations often contain additives that can cause bronchospasm or coughing, among other side effects.13 Moreover, their biochemical suitability for inhalation has often not been tested, and some parameters may worsen tolerability.11 Similarly, intravenous antibiotics are formulated with different salts, and some may be better tolerated than others, which requires clarification. For instance, colistin sulfate is used orally or topically, while the inactive prodrug sodium colistimethate is more suitable for injection or inhalation.9,14 Interestingly, most studies on tobramycin use the sulfate salt, but one small study suggests that the free base form may be better tolerated in terms of reducing coughing.15 Lastly, manipulating intravenous vials to reconstitute powders or dilute solutions increases the risk of contamination and potential infection.13

Therefore, intravenous formulations should be thoroughly studied for their osmolality, pH, ion concentrations, and the presence of additives that are not recommended for nebulization before they are administered to patients. This study aimed to assess how frequently the literature reports the preparation methods and biochemical parameters of inhaled antibiotic solutions through a systematic review. We also aim to explore the use of these solutions in our country and analyze their biochemical characteristics to identify which could be potentially safer and better tolerated.

The study was divided into three parts: a literature review, a hospital survey, and biochemical analysis.

The results of the review highlight the need for more reporting on the biochemical parameters related to the safety and tolerability of non-commercial dilutions of antibiotics for inhalation. Less than 10% of the studies identified in the review reported data on pH, osmolality, or relevant ion concentrations in non-standardized mixtures that patients must prepare at home. Moreover, it is clear that there is a lack of standardized protocols for preparing these solutions. To the best of our knowledge, we could not find reference texts in the literature that provide detailed instructions on preparing such solutions, despite their widespread use.

Regarding the hospitals surveyed, 67.4% reported using intravenous preparations for inhalation, including ampicillin, amikacin, aztreonam, cefotaxime, ceftazidime, gentamicin, imipenem, meropenem, tobramycin, and vancomycin. A notable issue is that some hospitals continue to prepare inhalation solutions from vials designed for parenteral use, even when commercial inhalation formulations exist, as is the case with aztreonam. The reasons may vary, including the lack of availability of these drugs in certain hospitals (due to cost, for instance16) or clinicians' preference for parenteral formulations based on their historical success and reluctance to adopt new formulations. However, it is important to remember that inhalation formulations have been tested in clinical trials and are specifically designed for efficacy and safety in this route of administration. Their biochemical parameters are optimized for inhalation, ensuring stability. For example, aztreonam has been co-formulated with lysine. Studies show that the pH for optimal compatibility between the powder and diluent should range from 4.0 to 8.2 (with maximum solubility and stability at pH 4.5–6.0), osmolality should be between 150 and 550 mOsm/kg, and permeant anion (e.g., chloride) concentrations should be kept to a minimum.17 The powder is more stable when refrigerated before reconstitution, which is why the commercial brand of aztreonam lysine is sold as a refrigerated powder, with the diluent containing minimal sodium chloride.

Our biochemical tests showed that not all mixtures fell within the established safety and tolerability ranges reported in the literature. In terms of pH (ideal range: 4–8),4 vancomycin and gentamicin mixtures were too acidic. The FDA has granted orphan drug status to AeroVanc™ (vancomycin hydrochloride inhalation powder) for treating pulmonary methicillin-resistant Staphylococcus aureus (MRSA) infections in CF patients.18 Although this drug is not approved in Europe, it addresses a crucial therapeutic need, and thus, off-label use of intravenous vials is common, compromising safety and tolerability. Vancomycin is also classified as a vesicant.19 In the case of gentamicin, the pH was also outside the recommended range; however, some authors propose broader pH margins (2.6–10), within which gentamicin would be acceptable.20 Furthermore, the greater tolerability of colistimethate observed in a recent study conducted at our centre, compared to nebulized intravenous ampicillin and gentamicin, may be partly explained by its more favorable biochemical profile, particularly its pH and osmolality, falling within the recommended tolerability ranges.21

Regarding osmolality, the ampicillin solution, with an osmolality exceeding 1700 mOsm/kg, is noteworthy as it falls far outside the recommended range. This drug is highly hyperosmolal, and moreover with a pH above 8. Hospitals in our study reported reconstituting 1000 mg vials with 4 mL of normal saline. In the review, we found four studies that reported the use of this drug for inhalation.22–25 In the study by Máiz et al., researchers reconstituted 1000 mg in 4 mL of water for injection, lowering the osmolality to 1250 mOsm/kg.22,23 The authors specified the brand used for this mixture, which was the same as in our study. Two other studies reported even lower osmolalities (around 650 mOsm/kg) because they used half the dose (500 mg) while maintaining the same volume.24,25 Based on our findings, reconstituting ampicillin powder with water for injection, rather than normal saline, to reduce osmolality would seem advisable for doses of 1000 mg. Two other antibiotics in our study, cefotaxime and ceftazidime, also showed even higher osmolality levels, exceeding the measurement capabilities of our tools, although only one hospital reported their use.

In terms of ion concentrations, various authors emphasize the importance of high anion concentrations,7 particularly chloride, to improve tolerability.6 For example, several manufacturers produce inhalation solutions of tobramycin containing 300 mg of the drug in 4 or 5 mL. According to our tests, the pH, osmolality, and chloride ion concentrations were within acceptable ranges. However, some hospitals still use pure intravenous vials without chloride. Other hospitals use the same vials but dilute them with normal saline, improving the ion balance and making the solution more tolerable. From this perspective, dilution is recommended, provided the drug's concentration is properly considered.

Finally, in terms of excipients, it is preferable to use antibiotics specifically formulated for nebulization.1 These solutions typically contain drugs dissolved in aqueous, isotonic solvents that may include tested preservatives to reduce microbial growth. Sodium chloride and other salts are commonly used to adjust osmolality, while acids and bases such as hydrochloric acid, sodium hydroxide, citric acid, or phosphates are used to adjust pH. It is noteworthy that all marketed inhalation preparations in our study contained excipients classified in Group 1 in the FDA’s Generally Recognized as Safe (GRAS) database. This means that there is no evidence to suggest a hazard to the public when these excipients are used at current levels.26 By contrast, intravenous antibiotics may contain additives designed to enhance chemical stability. Some excipients, such as sodium bisulfite and EDTA, are approved for inhalation, despite studies linking their use to side effects such as coughing and bronchospasm.27,28 In our study, intravenous formulations of gentamicin, tobramycin, and amikacin contained sulfites, and some brands of amikacin and tobramycin contained EDTA. Despite this, no increase in adverse drug reactions or deterioration in quality of life has been reported in our centre compared to patients using preservative-free formulations.11 Additionally, intravenous antibiotics may contain preservatives like phenol, which can cause airway irritation, coughing, and bronchoconstriction and is not recommended for inhalation.27,29 Phenol is also classified as a neurotoxin by the National Institute for Occupational Safety and Health.30 In our case, intravenous tobramycin contained this substance. Other intravenous formulations, such as vancomycin, ampicillin, or cefotaxime, contained no additives.

We must acknowledge several limitations in our study. First, the literature review could have been more extensive, as we could have reviewed other databases. Additionally, we focused exclusively on antibiotics used in our setting and excluded other commonly used substances such as antifungals.31 However, we believe that focusing on this topic allowed us to identify a significant proportion of relevant studies. Second, regarding the survey, the response rate was low, despite our efforts to contact hospitals via email and telephone. The reasons may vary (e.g., workload, lack of incentives, lack of prioritization). Third, the tests we conducted only measured pH, osmolality, and key ion concentrations. We did not assess factors such as viscosity, temperature, other chemical compounds, or inhalation devices, all of which are known to influence tolerability.32 Moreover, it is important to note that these biochemical parameters were measured under ideal conditions, whereas some delivery devices, particularly ultrasonic nebulizers, heat the solution, thereby increasing osmolality and ion concentrations.33 Finally, it should be noted that this study does not include certain antibiotics that, although not identified in the national survey, are used in specific clinical scenarios. One example is nebulized linezolid in transplant patients colonized by MRSA or VRE (vancomycin-resistant Enterococci). This absence may limit the generalizability of our findings to all possible clinical contexts.

Among the study's strengths, this review is, to our knowledge, the first to summarize the biochemical parameters of antibiotic solutions and provide recommendations based on accepted ranges in the literature. Furthermore, it addresses current clinical practices and offers alternatives for achieving better tolerability. We analyzed the biochemical parameters used in numerous hospitals, which strengthens the external validity and applicability of our findings.

In conclusion, more information should be reported in the literature on the procedures for preparing inhalation mixtures. Many hospitals use intravenous antibiotics for inhalation, and some of these solutions have biochemical parameters that fall outside the recommendations for optimal tolerability and safety. By considering pH, osmolality, chloride concentration, and the presence of certain excipients, we can better predict which solutions may be more tolerable for patients.

Manuel Vélez-Díaz-Pallarés: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. María Ángeles Parro-Martín: Supervision, Investigation. Hilario Martínez-Barros: Supervision, Investigation. Beatriz Montero-Llorente: Supervision, Investigation. Miriam Menacho-Román: Supervision, Investigation. Ana Gómez-Lozano: Supervision, Investigation. Rosa Nieto Royo: Visualization, Validation. Luis Máiz Carro: Visualization, Validation. Ana Álvarez-Díaz: Validation, Supervision.

Not applicable.

None.